MediaTek Inc

TWSE:2454

Earnings Call Transcript 2018-Q2

[Operator]

Welcome to the MediaTek 2018 Second Quarter Investors Conference Call. Your speakers today are David Ku, MediaTek CFO and spokesman; and Jessie Wang, MediaTek's Manager of Investor Relations. Ms. Jessie Wang will report first (sic) [ second ] quarter result, and Mr. David Ku will provide prepared remarks. And after that, we will open for Q&A.

Now I would like to turn the call over to Ms. Jessie Wang. Ms. Wang, please go ahead.

[Jessie Wang, executive]

Good afternoon, everyone. Welcome to MediaTek's Second Quarter 2018 Conference Call. As a reminder, all content provided on this teleconference is for informational purposes only, not intended for investment advice. Neither the issuer nor any of independent providers is liable for any actions taken in reliance on content contained herein.

MediaTek provides non-TIFRS financial measures as supplemental information. Earnings distribution is made in accordance with financial statements based on TIFRS. Unauthorized recording or redistribution of the video, audio, text and the presentation content of this teleconference is strictly prohibited. By participating in this teleconference, you agree to accept the foregoing terms and conditions.

Now let's start with the 2018 second quarter financial results. The currency here is in NT dollars. Revenue for the quarter was TWD 60.5 billion, up 21.8% sequentially and up 4.1% year-over-year. Gross margin for the quarter was 38.2%, down 0.2 percentage points sequentially and up 3.2 percentage points year-over-year. Operating expenses for the quarter were TWD 19 billion compared with TWD 17.2 billion in the previous quarter and TWD 18 billion in the same period last year. Operating income for the quarter was TWD 4.1 billion, up 112.1% sequentially and up 73.5% year-over-year. Operating margin for the quarter was 6.8% compared with 3.9% in the previous quarter and 4.1% in the same period last year. Net income for the quarter was TWD 7.5 billion compared with TWD 2.7 billion in the previous quarter and TWD 2.2 billion in the year-ago quarter.

Net profit margin for the quarter was 12.4% compared with 5.4% in the previous quarter and the 3.8% in the year-ago quarter. EPS for the quarter was TWD 4.75 compared with TWD 1.69 in the previous quarter and $1.51 in the same quarter last year. Please note that net income, net profit margin and EPS this quarter include one-off nonoperating disposal gain, which were not included in the previous quarter and the same period last year.

We also provide non-TIFRS financial measures, which excludes share-based compensation, amortization of acquisition-related assets and tax effect. Please refer to earnings press release and the presentation for details.

For the third quarter of 2018, we expect revenues to be in the range of TWD 62.3 billion to TWD 67.1 billion, up 3% to 11% sequentially at a forecast exchange rate of TWD 30.4 to USD 1. We are forecasting the gross margin at 38.2%, plus or minus 1.5 percentage points, and the quarterly operating expense ratio to be at 30%, plus or minus 2 percentage points.

For shipments, we expect shipments of smartphones together with tablets to be 100 million to 110 million units in the third quarter.

And now I would like to turn the call to CFO, Mr. David Ku, for prepared remarks.

[David Ku, executive]

Good afternoon, everyone. I think before we get into Q&A, I would like to spend some time to talk about or providing a few comments about our 3 major business products -- segments.

First one, we talk about the mobile computing. In general, the mobile computing, including smartphone and tablet, which account for 37% to 42% of our Q2 revenue. Smartphones, especially with the new product, Helio P60, have been quite successful in Q2, that's why you see that for Q2, our revenue is somehow slightly exceeds our original guidance. On top of the P60 -- successful P60 launching and also the engagement with the key customer, we also launched 2 main products in Q2, which included the Helio P22. And also, we'll have the Helio A22. P22 is slightly lower compared to P60 and A22 is slightly lower than P22. But in general, actually, right now the Helio conquered it pretty well both for the mainstream and also for the entry level. Also, I think one of the key design win we have on A22 -- Helio A22 is Xiaomi's new product, which brings margin in Q2 already.

From a shipment perspective, overall, for the mobile computing product, we shipped around 100 million to 110 million in Q2, which slightly exceeds the original guidance. When we're talking about the shipment, which including the smartphone and tablet shipment in total. Also, from a marketplace perspective, we still see the consolidation trend among the Tier 1s continued. So for this year, we see the Tier 1 guys or the big brand or the big China operators continue to consolidate the market shares. And in consequence, we see the second-tier brands or smaller brands turn somehow so conservative in the second half, which is part of the reason we see [ surging ] last seasonal growth are in third quarter.

On the product portfolio side, in addition to the P60, we also expect to launch the new product, which is equivalent of our Category 12 and also Category 16's modem capability SoC before end of this year. We do believe with the new product coming out before end of this year, we will continue to extend our product portfolio and also further enhance our technology [ capacity ] into the next-generation products.

The last one of these for the smartphone products as we announced our 5G plan in -- actually that's around June this year. I think just want to reassure everyone that we are actually one of the leaders for 5G product portfolio. And based on our current portfolio, we are fully ready for commercial launch for 2020 5G market. The first Helio M70 models will be coming around next year. And also, the first set of 5G will be coming around by end of next year, end of 2019 and also beginning of 2020 as well. But in general, I think our product portfolio is fully ready for a big pickup for 2020 for 5G especially. I think that concludes about the mobile computing business update.

Let's move on to the next sector, which is the growth sectors. Growth sectors, including IoT; PMIC, power management IC; and also ASIC. For these quarters, the growth sectors account for 25% to 30% of our overall revenue. We see the 7 quarters' revenue of growth sectors on a quarter-over-quarter basis actually goes in line with our whole company growth. But for third quarter, we see a much stronger growth compared to the whole company growth in the third quarter, [ 3% ] for the third quarter and always pretty strong seasonality for growth sectors. Especially, I think for the IoT market, we see a fairly strong pickup for the VAD, which stand for Voice Activated Device, pickups in China. And also, we have a new product -- I mean for AI, audio-integrated device. But in general, I think we see pretty strong growth for the growth sectors in third quarter this year.

In auto and IoT product, I think PMIC, power management IC, continue to grow our revenue in multiple platforms, which included smartphone and also VAD.

The last one in the growth sector is really our ASIC business. I think right now, ASIC business is mainly coming out from the consumer products' revenue -- consumer product segment. But right now, we have a pretty good design in into the networking product. And based on the current design in and design win and also takeout situation is often pretty smooth and successful. We do expect we will see some networking product revenues hopefully coming out, starting from next year in 2019. That conclude the update on the growth sectors.

The next one is with our hardware product, which including digital TV, feature phone, optical storage and DVD. For the hardware product sector in Q2 revenue-wise, which account for around 25% to 30% of our overall revenue, I think the growth rate in third quarter in hardware product is pretty healthy and is actually stronger, is higher than corporate average, mainly driven by signal demand. Especially, I think, for the digital TV, right now, we're in the progress -- in the process of our fully integrated MediaTek DTV team and MStar DTV team.

We do expect the consolidation plans will have some synergy coming out next year. And I think overall both from the operating expense perspective and also, hopefully, from the market share perspective, we will see some further improvement next year after we realize our operating synergies. I think that's a quick update for all 3 product segments.

In the end, I think, before getting into Q&A, we also try to highlight, due to the relatively weak third quarter's seasonality, right now, for the full year perspective, we're trying to expect the full year revenue, basically 2018 versus 2017, the full year revenue mainly just coming down a little bit year-over-year basis, I would say very low single digit down, so you can consider the financials come down a little bit. But I think as explained earlier this year, the focus of this year is really trying to have a profitable growth. So if you focus on the gross margin, operating margin, and more importantly, operating margin dollars, you will see pretty strong year-over-year growth for this year. I think that's the overall strategy for this year.

On top of the gross margin and also operating margins improvement, we are also, when we talk about profitable growth, we are also committed to looking for diversification for a blended digital growth across multiple platforms. So mobile computing this year will account roughly 35%, 40%. Growth sectors will come for another 30%. And also, the cash [ conversion ] is on the hardware business. I think the strategy is we will be trying to reallocate the resource among our tiers without increasing the overall resource but trying to invest more resource with growth sectors and also we [ serve ] with our 5G investment. So we do believe that it won't be too long for us and we should be able to create a much more balanced growth portfolio, on top of our getting healthy or recovery business on the smartphone side. In the meantime, I think having a profitable growth, especially on the gross margin and also operating margin side, is also a very clear evidence to see the first stage go with the kind of guidance fulfilled.

I think that conclude my quick update, and we can get into Q&A directly.

[Jessie Wang, executive]

Thank you, David. We are now ready for Q&A session. May we have the first question, please?

[Operator]

[Operator Instructions] First, we are having Randy Abrams from Crédit Suisse.

[Randy Abrams, analyst]

The first question I had, maybe 2 clarifications from the Chinese conference. First, on the ASP trend, I think you made a comment about the Helio ASPs. But if you could give a sense of the overall smartphone ASPs for both second quarter and then outlook for third quarter. And then the other clarification was on seasonality for the mobile products in fourth quarter. I guess your expectation at this stage, I think you're expecting the growth, but maybe what's driving that, whether it's new product launch or market share or a view on the market?

[David Ku, executive]

Okay. Randy, thanks for asking these clarifications. I think for -- during the Chinese call, when we talk about ASP, we're actually talking about the blended ASP, not just the Helio ASP, first of all. And also, when we are talking about it's going to be 0% to 5% down, we're actually talking about the third quarter's. Because after the Chinese call, some people have called and asked about the second quarter. I think for the second quarter, due to the higher P60 contribution, for the second quarter, ASP is actually coming down -- up a bit. So I would say for Q2, blended ASP, due to higher P60, is overall will be 0% to 5% up, up for Q3, because we see a more sort of P22 and also A22 coming up. So the ASP, the blended smartphone ASP, will come down a little bit, also 0% to 5% down. I think that's on the ASP, smartphone ASP front, that's clarification number one. I think for the second question, talking about the third quarter's growth, I think overall, third quarter right now for the whole company, I think we're looking for 3% to 11% quarter-over-quarter growth. Among 3 different product segments, I will say both growth sectors and also the hardware sectors have higher growth rates than the corporate average. I think for the smartphone, its growth rate is relatively weaker compared to the corporate growth rate.

[Randy Abrams, analyst]

Okay. David, I wanted to also ask, I think you made a comment also about fourth quarter, but if you could clarify your expectation, whether fourth quarter you expect mobile to grow and what may drive that growth for fourth quarter?

[David Ku, executive]

Okay. I think for the fourth quarter, I can't -- we're not -- figuratively speaking, we'll not be able to provide fourth quarter guidance. So during the Chinese call, we're kind of talking about -- we're only commenting about from a market perspective or market demand perspective. We do believe, right now, based on what we see, we should be able to see relatively stronger quarters, basically, the growth quarters on the smartphone side, the fourth quarter versus the third quarter based on the visibility or data point we have right now. I would say it's mainly due to, I think, 2 reasons. First of all, for P22 and also A22, even though we stopped the launch end of second quarter, beginning of summer, actually the mid of third quarter, I think sometimes it takes some time to ramp it up. So the new design in, we will see more ramping up in fourth quarter. I think that's reason number one. Reason number two is actually, soon, we'll have 1 or 2 new products coming out like in the second half this year, and hopefully, that will just further fuel up the growth on the smartphone side.

[Randy Abrams, analyst]

Okay, great. The second question on the margins. You mentioned the target to still get to 40% as a medium-term goal. Could you talk, I guess, about the 2 elements? One, within mobile, with the cost down, I think, reaching -- or the new architecture about 70% by year-end. I guess your expectation from their mobile before 5G, if it's more stability, and to get to 40%, it's more from the nonmobile products?

[David Ku, executive]

Well, I think for the gross margin -- for the gross margin goal, if you like, like the CEO, Rick, talked about during the Chinese call, we're still aiming for 40% gross margin. And also, I think, to be precise, it should be 47%-plus and also 10%-plus as the long-term -- mid- to long-term goal as our profitability. Given the current gross margin profile, I think, for the second quarter will be like 38.2%. For the third quarter, I would say pretty much within this range of, plus or minus 1%. It really depends on the -- at any given quarter, right. It really depends on the BG or business group mix. And also, on top of that, I guess, fundamentally, we're still looking -- for next year, I guess, we're looking for an opportunity to further grow our gross margins on the smartphone side. I think that's the plan. That's the strategy. But in terms of the detailed strategy, I think it would be a combination of the different or the higher product mix, like during the Chinese call, we talked about for next year for Helios development on third quarter, we have Helio P22 and also Helio A22, which is all, relatively speaking, actually at a lower segment compared to P60. But next year, I think we have a new product coming out as the higher segment compared to P60. So hopefully, the patterns or the higher subproduct segmentations will share some -- help on the gross margin side. I mean, that's one of the reasons. Another reason around the strategy is we will continue to improve our cost elements or especially the cost architectures on the smartphone side. I think that will be an ongoing pursuit.

[Randy Abrams, analyst]

Okay, great. One feature, I think, you've been marketing at some of the conferences that the 3D sensing works more at lower-cost solution. Could you maybe talk about your potential outlook for that and if it's an angle you see in terms of gaining either ASP or market share through offering that platform?

[David Ku, executive]

Well, I think to be precise, it actually results for the quarter is from 2D or 2.5D sensing. The general idea is actually we'll use 2 camera, 2 strong camera, which are performing the very similar function of 3G sensing. From [ core data ] perspective, the benefit of that is actually it will have, I will say, extends 3D sensing capability but at a much lower power cost charging. So far, I think we have 2 customers started out in that. And we don't know if there'll be kind of major features still. We have time to see the final market demand. But just from a performance perspective, we feel very comfortable of our solution, and hopefully, that would lead to a slightly higher ASP or a lift movement on the current ASP. I think that's up with the 2D or 2.5D's recognition functionality.

[Randy Abrams, analyst]

Okay. And one final quick question. Your inventory ended at, I think, 90 days off of ending balance. If you could give a view, expectation into second half, where you see your inventory trending.

[David Ku, executive]

I think for Q2, the days of inventory is around 86 days. For Q3, I think we will pretty much maintain the same level of days inventory. I would say in the range of maybe like 85 to 100 days, maybe that's at a wider range. But I think, most likely, it probably will be very similar to Q2.

[Operator]

Next, we're having Stefan Chang from Maybank.

[Stefan Chang, analyst]

Just 3 quick one. The first is also about some clarification. So I remember in the Chinese call, you also mentioned about the 4-year smartphone shipments could be up slightly year-over-year. And it's based on the last year number or the first half number as well as Q3 guidance. I think that implies that Q4 could be up by at least 10% to 20% sequentially. I just wonder if the take is correct.

And actually, also a follow-up on this. Because if the smartphone implied Q4 seasonality like this, but based on the 4-year revenue strategic kind of guidance, I think Q4 revenue could be roughly similar to Q3 midpoint. So does that mean the other business will go down in Q4? So this is my first question.

[David Ku, executive]

Well, first of all, right now, we will not give out exactly the guidance for Q4 on the whole company perspective, so just please bear with us. But just trend-wise, I think for the smartphones in Q1 -- I mean in Q1 on year-on-year basis, I think shipments are actually coming down. For Q2 and Q3 are actually pretty flattish if you compare it on a Y-o-Y basis. Q4 again is -- on 2018, based on the current visibility, we do foresee volume -- both from volume and revenue perspective, I think for smartphone, not only for the smartphone, but we actually have a pretty good opportunity to see both of them up. So I think that's probably the only measure we can provide for the time being. For other sectors, in general, fourth quarter -- as I explained earlier, for third quarter, in general, it's a pretty strong quarter for the growth sectors and also for the hardware sectors. But for Q4, in general, that's -- there's a normal seasonality down for growth sectors and also hardware sectors.

[Stefan Chang, analyst]

Okay. This is very clear. And also, the second question is about the smartphones -- or the follow-up on smartphone ASP. I think based on the released number and also the sales mix, it looks like this year, so far, the smartphone ASP is declining year-over-year. And I just wonder if this is due more to the product mix or because of the competition. And if you look at the next year, I understand you mentioned about some even higher end product. But how do you view the combined impact from the competition as well as the product mix and the impact to the ASP in the next year, if you can give any broad color?

[David Ku, executive]

Well, I think for the full year blended ASP, first of all, we start on full year blended ASP. I think for this year versus last year, it should be pretty flattish. For example, this year, we see a lot of P60 -- growth in the higher-end products coming out. So the ASP should be flattish -- pretty flattish and maybe even up a little bit, especially judging from both from the Q1 and Q2 ASP, if you recall, we give out -- we just gave out a guidance -- with our guidance, we just give an update for the ASP though multiple platform. But remember, for Q1, actually, we provided similar guidance. So in this first half, always on a Q-on-Q perspective, it's actually up a little bit. So I guess when you clarify actually for 2018, our fully year ASP should be pretty flattish. It's up a little bit.

For next year, it really will depends on how successful we can continue to push for the high-end product, which means the high -- entire thing, the P60. But right now, it's a little bit too early to tell. But the general the idea is to, somehow link that to the earlier questions asked by Randy, the ASP, so we do -- from a product portfolio trending perspective, we do try to introduce a much more balanced and I will say, high-end focus will be much more balanced [ front of a wheel ], which also including the new segments, which is higher than P60. So -- but we also need to think about the overall competition landscape next year. But right now, it's going to be a little bit too early to be commenting on that. But the general idea is it will be a new product higher than P60 next year and continue to enhance on the entry level. And hopefully, by adding new features, we can somehow either have a flattish on the entry level and also mainstream or maybe just a slow decline ASP on the entry level and also mainstream.

[Stefan Chang, analyst]

Yes, understood. I guess, last year, [ base ] probably altered by the Q1, but I understand where you're coming from and that is very clear. And very one last quick question is about, I guess, on operating margin. I think the company has already done a very good job in gross margin improvement. But so far, operating margins look still volatile, probably due to the revenue scale. So I was just wondering, when you mentioned about the midterm goal for the gross margin and operating margin, what is the time frame do you define the midterm?

[David Ku, executive]

We don't have a fixed time frame, but hopefully, when we say midterm normally, hopefully, it's actually a 2-year profit gain.

[Operator]

Right now, we're having Gokul Hariharan from JPMorgan.

[Gokul Hariharan, analyst]

First of all, could you talk a little bit about the ASIC business? Can you talk and give any details about how does the ASIC business is looking like? You've mentioned that you're starting to have some success beyond consumer and networking. Can you talk a little bit about what kind of wins are you having? And what does the pipeline look like beyond networking? Do you have engagements in more data center, AI-related stuff or it's still going to be mostly consumer-centric kind of projects?

And maybe broadly, you could talk a little bit about what are the criteria that you are applying in terms of location in ASIC as well as criteria to pick up some of these projects, given, I think, there seems to be a lot of ASIC projects coming to the market as well as to you guys?

[David Ku, executive]

Well, I think, first of all, before I get into any detail with ASIC business, one thing to kind of remind actually peers before lots of ASIC questions, most of our ASIC customers, they are all global Tier 1s. So by the service contract, we will not be able to provide any customer-specific information. So I can only be providing the general sort of the rationalized information. So well, first of all, currently, most of the revenue, the ASIC revenue, are pretty much all coming up from the consumer product, mainly basically the game console-related product and also a little bit from the multimedia-related product. I think that's majority of our revenue. Revenue of it actually both for the game consoles and also for the multimedia, which including VR, so -- and AR, to some extent, are all still growing. So the overall strategy for our ASIC business is we're trying to further enhance our growth and also revenue size based on the consumer product, again, which including but not limited to the game consoles, AR, VR and some multimedia product. And so for this year, growth rate looks solid. For next years, we still believe we have a growth opportunity, even based on consumer products. So that's strategy number one.

Strategy number two is currently based on the design in, design win and also the engineering development, we're actually getting into a new sector, which is called the data switch. This is a high-end and very successful service technology. I think maybe with what we just announced -- we're probably worldwide #1 to officially announce service technologies [indiscernible]. We believe we are probably one of the leaders on that front. And with that technology, I think we can focus on the -- we will probably start with the data switch business first. But when you think about high-speed service, I think they are up. Also, in other location, there will be some other area where we'll explore opportunities as well. But the problem for the ASIC business is it sometimes takes the design in and the design wins back is pretty long. Normally, the design in -- design win versus the revenue coming up is, in general, I think about 1-year plus, sometimes actually 1.5 years to 2 years. So even we have any plans looking for beyond the networking switch business, I guess, we probably still need to wait to decide -- if you're coming from now, I would say probably 2-plus years, but we do have plan to base on the similar IP to looking beyond the network switch business.

[Gokul Hariharan, analyst]

Okay. Could you talk a little bit about any quantifiable numbers in terms of number of design wins or something like that? I think I understand that customer-specific revenue numbers are quite sensitive, given it's a custom product. But could you talk about the number of engagements that you have and how does this change over the last 12 to 18 months or something like that?

[David Ku, executive]

Unfortunately, we are -- probably, we will not able to provide any more color on that.

[Gokul Hariharan, analyst]

Okay. No worries. So just sticking with the growth engine segment, David. I think in the Mandarin call, you guys highlighted ASIC voice assisted devices in China, NB-IoT and probably PMIC as the key areas. Could you talk a little bit about where is the growth likely to be the strongest in these categories? Is it more at the consumer IoT like NB-IoT, VAD kind of products where the growth is going to be strongest? Or you think ASIC is also going to kind of come in as a big growth driver going into next year?

[David Ku, executive]

In terms of growth rate, I would say NB-IoT is probably one with simply the strongest growth rate because right now, NB-IoT base is very low. By looking to the second half this year and also with the visibility we have on next year's, I will say it's probably the one carrying with the strongest growth rate. And on the absolute scale, I think NB-IoT this year is still relatively small. But for next year, we do kind of expect NB-IoT will become pretty -- become one of the meaningful revenue on NB-IoT, but in the growth rate, I think they're the strongest. The growth rate, I guess, if you focus on WiFi, focus on VAD, voice assisted device, and also focus on [ entry level ] and I will say the growth rate is somewhat similar. Maybe VAD -- relatively speaking, the VAD is higher [indiscernible] we see pretty strong pickup. VAD, I think last year was in U.S. and Europe. This year is in China. I think relatively speaking, VAD probably has the higher growth rate. But WiFi looks good and also machine-to-machine and also some type of [ devices ] looks pretty solid in terms of growth rate.

[Gokul Hariharan, analyst]

Okay. Understood. So when you talk about your 5G plan, I mean, if I jog my memory and recall correctly, I think for 4G, your stand-alone modem and your SoC pretty much came out around the same time and we didn't hear much about the stand-alone modem. 5G looks like you're going with the SLIP modem next year and SoC following [ M320 ]. Is there a reason for the change in strategy in terms of the timing gap? Is there a clear -- is there credible SLIM modem business that -- to be had either in mobile or in other vendor end markets given that, historically, you guys have not really had a meaningful SIM modem business?

[David Ku, executive]

Well, first of all, I guess, we're not -- currently, we're not looking for -- we're not looking aggressively for a SLIP modem business. So the SIM modem is what we -- I think the overall strategy about having a SIM modem is first, is when we have the SoC just like what we did for 4G is really for the IoT task, basically the carrier certification because every time we have a new generation of modem coming out, so it will be relative easy and faster to certify SIM modems. Because after all, the SoC, just the SIM modem plus AP and we feel fairly comfortable of our AP technology. But after the SIM modem, especially for 4G to 5G, I guess given the fact right now that even this sort of the industry standards giving the moving target, it will be better to have a SIM modem first to work closely with the equipment vendor, the base-station equipment vendors, plus the carrier, then comes IoT. I think that's a similar strategy. And -- but I guess, my point is, we're not trying to use that same modem to aggressively looking for SIM modem business, okay? So that's point number one.

Point number two, I think based on your description, it sounds like we have different strategy for 5G versus 4G. I would say yes and no. Basically, the major difference is when you think about for 4G versus 5G, for 4G, by the time we have 4G, regardless of the SIM modem for SoC, from an absolute industry perspective, we are probably -- some people say a year, some people say 2 years behind overall industry development. But for 5G, I guess we're among the leader group because for next years, even from the operator perspective, next year will be the precommercial launch of 5G, which means the schedule [indiscernible] a huge 5G smartphone picking up. You will see a few be coming now, but it's not going to be in mainstream. Overall, even the most aggressive view, we're talking about 2020, you will see some, I would say meaningful effect, mainly for small balance coming out in 2020. But our overall profit for -- is aiming for that, especially ready for that. So for me, that's probably the major difference when you're talking about 4G versus 5G. To make the long story short, I think 5G, we are ahead of the curve. We are much more ready for the first wave -- battle for the 5G. And also, actually, the 4G, the 5G modem -- standalone modem versus SoC, is really coming out in the same year because we're talking about end of next year, beginning of 2020, we're really commenting from the cost manufacturing perspective, okay, not from a [ technology ] perspective. I mean, [ technology ] normally is like 6 months ahead of customer shipping. So it's really just the time frame. The only difference is that one will have standalone modem, we can just -- much bigger complete of the IoT test, the carrier certification. I think that's the situation.

[Gokul Hariharan, analyst]

Okay, understood. So one related question slightly, David. So your competitor, Qualcomm, has been talking about prepaid 5G RF solutions kind of like bundling like the RF with baseband processor for 5G to reduce the potential design time complications. Any plan -- any thoughts on how MediaTek is going to address this? Is there a closer partnership in terms of the existing RF vendors, that MediaTek think about getting into potentially RF? I think you guys have some initiatives out there of smartphone, but is there any thoughts on that?

[David Ku, executive]

Currently, I think RF is actually not our focus -- not our product focus because, right now, I think the focus really is just getting 4G out here, we're trying to launch a new product portfolio. In the meantime, we're trying to accelerate 5G development. On top of mobile device, I guess, we're also trying to allocate or reallocate some of the resource by getting to the growth sectors, which including IoT, the PMIC, the ASIC. I mean, that's the overall strategy. And bear in mind, like we're explaining and trying to point is, with all new products, new technologies have been coming out from MediaTek, the overall R&D resource have become [indiscernible] so we really didn't reach that. So we really need to prioritize and also strategize about which area we want to go to, which area we will not get into. So for the [ RF ] area we decided not get into at current stage, given the overall consideration. But on the other hand, we do work very closely with our [ OSH ] partners and basically just the 2 or 3 major [ OSH ] partners out there, we're all working very closely both for 4G and 5G [ RF front-end ]. And we really don't see that as a disadvantage for us, for more 4G and 5G. Because after all, [ RF front-end ] we do believe actually it's probably the best strategy given our current situation to work with external vendor really, than try to do everything else.

[Operator]

Next to ask question, Brett Simpson, Arete.

[Brett Simpson, analyst]

David, can you just maybe talk a bit about the entry segment of the mobile phone market, of the smartphone market? And I just wanted to get your sense for the competitive environment here, because I mean, it seems like Qualcomm is not a successful player in entry. They're not a big player, not part of the market. And I think it's well-known that Spreadtrum has been struggling. So I mean, it sounds like you must have a very small market position in this segment of the market. I'm just trying to understand what other scope to see a much higher gross margin returns as you are able to sort of leverage more of the position you have on that segment of the market?

[David Ku, executive]

I think, so far, for the entry segment, in general, I think our market share [indiscernible] in the segment. And from a gross margin perspective, I think, overall, we feel comfortable. If I can say -- I will say definitely the market position is really just, right now, we have a much better cost structure product. So overall, I think the gross margins among all segment, which includes entry-level and also the mainstreams are overall getting better, all getting better. And given the customer's needs, I should say the customer's full needs for entry levels, I guess we do our turnkey solution and also our huge operation in China does give us some unique advantage in supporting this group of customers. We will definitely try to advance that going forward. But if you take one step back, when you think about the smartphone competitive landscape perspective, I'm not talking but the chipset perspective, I'm talking about from phone perspective, there's another big trend, which is the big brand is actually getting more and more market shares. So basically top 5 or top 6 brands, you name it, are just getting more and more market shares every year. So we also need to take that into consideration as well because when the big brands are trying to get into taking more market share, chances are they're trying to, based from their perspective, they also thinking in a way to increase their phone ASP.

So our chances are they're just promoting more managing products. And -- but in general, I think from our perspective, that should be a good news because right now, especially for this year, for the first half this years, our market share for the top 5, our top 6 brands are all getting up meaningfully. And so if the overall market trend is actually the top-tier guys, it just consolidating more market share. And on top of that, the top-tier guys trying to sell more mainstream products or mainstream-type products. I think that should be a positive trend, from a MediaTek perspective, both from the ASP perspective and also from the overall revenue perspective. I think that's the update.

[Brett Simpson, analyst]

Okay. And just in terms of the consolidation process, can you just give us a sense from where you are -- where we are in that process? The top 4 have been clearly structurally taking share, but what portion of the market as the top 4, sort of, taking new TAM today and where was it a couple of years ago?

[David Ku, executive]

I think for last year, we know what top 6, top 7, but the top 6, top 7, I would say probably account for a good 70% to 80% of overall market share from a smartphone perspective last year. This year, I will say, it's similar, but it's definitely getting higher, I would say like 80% plus. I think there is always this and this. And both for China -- when I say the market share, I'm talking about basically much of China, because really China is an emerging market.

[Brett Simpson, analyst]

Got it. And just in terms of your ASP as a percentage of the selling price of phones in China, I mean, we're at record low levels today. I mean, we haven't seen -- I mean, we're structurally seeing some of the ASPs go up, and we're seeing your ASPs fall. And I'm just wondering, do we see this situation reverse at some point? Is this the new normal where you will continue to see your ASP as a percentage of your customer's handset price continued to fall? Or how should we think about that?

[David Ku, executive]

I would say -- probably, to sort of say in terms of ASP -- I should say, in terms chipset value or chipset ASP versus the overall [ wallet ] cost, basically is your question. I would say it's actually staying up the same, holding up, okay? And so again, in general, but if you're looking at the top 5 or top 6 guy, that one is actually holding up pretty well. Sometimes, they even keep going up because, as I've explained earlier, they're also trying to upgrade their product portfolio. And chances are getting a mid- to high-end SoC, it's trying to important factors for the phone maker to upgrade their product, upgrade the ASP. So for Tier 1, actually holding up quite well, maybe even with the opportunity of going up.

For the other players, mainly more of the entry-level folks players, I will say these maintain, also probably down a bit because most players actually are competing very aggressively with the goal of the top 6 -- or top 5, top 6 guys, so they need to be aggressive on pricing. So in general, the turnaround pushed down basically everything, online [ phone ] cost.

[Brett Simpson, analyst]

And then maybe just a question on OpEx. I mean, it's growing faster than sales growing on a year-on-year basis. And I'm trying to get a sense, at what portion of your OpEx has been mobile today? Particularly if it's 5G now under development, what portion of OpEx is smartphone?

[David Ku, executive]

I would say it's pretty much in line with the [indiscernible] contribution. Smartphones right now account for 40% of our overall revenue, right, even though right now, there's no revenue for 5G. But for 4G plus 5G, the overall resource we spend is, accounting-wise, actually pretty much in line with the overall 4G resource right now we have.

[Brett Simpson, analyst]

Right. Okay. And then maybe just last question. David, your name is always mentioned as a beneficiary of AI, and we're starting to see in your Helio series both the A series and P series, we're seeing AI as a feature more and more in your portfolio. How should we think about the revenue opportunity for MediaTek? Today, it seems like it's embedded inside the [ dial ] of the SoC and so there's no specific revenue that you can attribute to AI. But how should we think about the, sort of, requirement that's going forward and some of the other segments you play in, like TD? How should we think about AI as an opportunity for MediaTek over the long haul?

[David Ku, executive]

I think from a revenue perspective -- maybe it's either I'll why or explain that concept. From the revenue perspective, I mean, AI brings 2 opportunities. The first one is, really, we see AI as an enhanced features. Maybe one idea is just comparatively small, like in the past we had VGA all the way to upgrades like SD and even 4K. AI, basically from our perspective, for a lot of products, including but not limited to smartphones and also TD, will become the new features. So from an AI in terms of revenue impacts, with new features, sometimes, it's giving even higher ASP or better ASP maintenance. So that's one opportunity for AI for now. But combining AI, which is in product portfolio. Well, in otherwise, we are actually working on this right now, but currently there is no revenue yet. It's really just a -- AI as a new revenue stream, okay?

Arguably, I think it's all AI as new applications. I think, for example, for the VAD, [ for its new ] AI application, even though for the VAD, we only have half of the AI, not the full AI. But because of that new AI features, we can just somehow link that, just melding the AP functions into the VAD [ receiver ]. So again, I think the quick summary of this is, in general, I think we do see AI as a positive opportunity for MediaTek, both for the existing business perspective and also for new business perspective. The only problem is the existing business in [ dual-core ] AI is relatively much faster because, after all, nobody will get the idea about better functionality of products. But for AI, the brand new revenue for some of these [indiscernible] materially low revenue, I think that's something -- there's a product we're still working on right now, but currently, we don't have any revenue. That's still only [indiscernible].

[Brett Simpson, analyst]

And do you think, just on the point about ASPs declining in mobile, when you look at 5G and when you look at where AI is going, do you think these are technologies that will absolutely reverse your ASP decline? It may be that ASPs sort of probably start to rise meaningfully because the penetration of these technologies becomes much more at scale. How should we think about the ASPs, particularly in your smartphone business over the medium term?

[David Ku, executive]

I would say for the smartphone business, the biggest drivers both on the revenue and also from the ASP regression perspective, is really 5G. For AI, I would say, it's more of a function enhancement, and were helped, from the ASP perspective but not a big time, no way. For 4G to 5G, you'll see pretty big jump from the ASP perspective.

[Operator]

Right now, we're having Callum Hughes, Indus.

[Callum de Courcy Hughes, analyst]

Just going on what you were saying about 4G to 5G, so one would be the ASP going up. But it seems that you're going to be in a much more competitive position than you were before. Would you say your market share is more likely to be higher than it is for 4G?

[David Ku, executive]

On a like-for-like comparison, the initial 5G versus the initial 4G, we do believe our initial 5G market share should be higher than our initial stage of 4G that was like 4 -- roughly 4 years ago.

[Callum de Courcy Hughes, analyst]

What about 5G versus current 4G?

[David Ku, executive]

5G versus the current 4G, we also believe we have opportunity to continue to grow the market share in general. So basically, I guess, when we talk about 5G versus 4G, probably the better way to think about that is just our overall market share of smartphone. We do believe with 5G coming out, the overall market share of smartphone, we still have opportunity to continue to grow.

[Callum de Courcy Hughes, analyst]

Okay, cool. And now for your new modem, you were saying 70% by the end of the year. Is that for the number of units? Or is that the revenue for mobile, 70% have been the new modem?

[David Ku, executive]

Units -- revenue. Sorry. Revenue, yes.

[Callum de Courcy Hughes, analyst]

Okay, and then -- sorry, just got one more. Did ZTE, did that end up having any effect on you?

[David Ku, executive]

I'm sorry, say that again. I didn't understand your question.

[Callum de Courcy Hughes, analyst]

The ban on ZTE, did that impact you at all?

[David Ku, executive]

I think ZTE has no impact on us, especially right now, they pretty much dropped off [indiscernible].

[Operator]

The next question is coming from Michael Chou, Deutsche Bank.

[Michael Chou, analyst]

David, just a follow-up question. Is it fair to say your smartphone gross margin should be flat to improved slightly in -- quarter-on-quarter in Q3?

[David Ku, executive]

Slightly -- up slightly, yes.

[Michael Chou, analyst]

Yes. Because you mentioned your new modem accounts for more than 70% of the total smartphone shipment by year-end. So is that fair to say your smartphone shipment gross margin could continue to improve quarter-on-quarter in Q4 this year?

[David Ku, executive]

Yes and no. Because, again, on a quarter-over-quarter basis, also one of the major functions, it's really just a product mix. If you think really about for Q1 and Q2, we continue to ship more P60. So that actually is a big supporter to the gross margin enhancement. But starting from Q3 and Q4, especially for Q3, like earlier explanations, we'll start to ship more P22 and also A22. That will offset a little bit of our gross margin enhancement. But for 4Q, it will really depends on what the segmentation mix.

But let me take one step back to explain about the gross margin. I guess, probably the better way to think about that actually is, as long as we can continue to enhance our cost structures, in general, we definitely have opportunity to continue to enhance the gross margin. But on a quarter-over-quarter basis, there are still a lot of other factors you need to consider. And based on the current visibility, both from a 3G and 4G -- 3 (sic) [ third ] quarter and fourth quarters, probably the better way -- either way you think about it, that is actually flattish to slightly up, maybe that's a better way to think about that.

[Operator]

Next one, we're having Charlie Chan, Morgan Stanley.

[Charlie Chan, analyst]

So I already learned a lot from the previous discussion, that was very helpful. So just some clarification. So first of all, in 2Q, you booked around TWD 3.6 billion to TWD 3.8 billion, came from auto shipment sales, right? But I thought you mentioned you should be TWD 5 billion, right? So will there be any [indiscernible] amount to be recognized in the following quarters?

[David Ku, executive]

Because that's exactly how much we're going to recognize every year, it would also depends on the business situation, so some of that will be pushed to the next year, you see.

[Charlie Chan, analyst]

Okay. So can quantify how much that would be for next year, David?

[David Ku, executive]

I probably won't be able to quantify right now, so I would say, probably we'll just provide a number by end of this year because it will be much more clear to us. But in general, with the [indiscernible] is just coming out next year.

[Charlie Chan, analyst]

Right. And I think it was also mentioned that several new features are for [ tempo ] K16 kind of a new product that would compete with [indiscernible] this year. So do I understand this right? You will have a new product, there will be K15, maybe 7-nanometer? And that is going to be competing with the [ 700M ]? Can you double confirm that the spec is right?

[David Ku, executive]

Let's confirm one by one. First of all, we will have a Cat 12 and Cat 16 product before the end of this year, that's confirmed. Okay. And confirm also that we will have product coming now in the 7-nanometer process, but that doesn't necessarily mean the 12-inch -- Cat 12 and Cat 15 will be a 7.

[Charlie Chan, analyst]

Okay. So they could separate products?

[David Ku, executive]

Right. Right.

[Charlie Chan, analyst]

Okay. And you also mentioned the potential [indiscernible] contents upside for next year, high-end product. So if it is not for AI, what other key features that you want to enable for those higher-end smartphone, by those new chip?

[David Ku, executive]

I think it's modem-wise, I didn't say modems were coming down. I should say, some modems is working. And also definitely AP including processors, another, and plus AI. So I guess, right now, we're talking about smartphone between 3 factors or 3 [indiscernible] Just basically modem certification, processor, which includes CPU and GPU and plus AI. So if everything goes well, we should be able to basically operate all 3 products.

[Charlie Chan, analyst]

Okay. And there is some industry development regarding -- you have seen the CD-ROM safe codes, those are core data. So you can see the, kind of, big demand for company like RITEK, right? And your company is find a [ controversy ] for those optical storage. Do you see any upside in this business unit?

[David Ku, executive]

I think from volume-wise -- well, first of all, I don't believe we're involved directly, so in the supply chain yet, based our DVD players and all the real players are actually still on the traditional consumer front, really on the corporate front, okay. Secondly, I guess even with the [indiscernible] I think the overall volume right now is still relatively small.

[Charlie Chan, analyst]

Okay. Yes, so lastly, regarding the new modem mix. Can you confirm this has driven mix or shipment mix, the 70% mix?

[David Ku, executive]

It is revenue mix, revenue mix.

[Charlie Chan, analyst]

Oh, revenue mix. So I'm just wondering, because if you look at all your new products on the roll mat, P60, P22, A22, even the low end 4G, MT6739, those are the new modems, right? So do you have any modem inventory on hand now that you want to sell in fourth quarter?

[David Ku, executive]

I think we still have some. If everything went well, most likely it's in -- slightly more in the quarters, actually [indiscernible] out of those, yes, it will be also up.

[Charlie Chan, analyst]

Yes. I just want to get a sense whether that new modem tailwind can sustain in to first quarter this year, right? Because I thought the new modem would be maybe 90% or 95% of your revenue in the end of this year. So just wondering how you're going to further improve your cost structure? Can you give us how to understand that?

[David Ku, executive]

I think for 2 things. First of all, for next year, modems compares with last year, right? But for next year, in comparison to, we will have a new product coming out. And not necessarily we would upgrade our overhaul the modem architecture, because this year, by and large, last year, we need to overhaul architecture because we need to have a bit saving on the modem side. But we're pretty much getting there already, we are seeing good result. But doesn't mean that no further room. I think further improvement will be much smaller, but we'll continue to improve that. So point number one, even for the modem itself, even though some of new modem versus the overall largest modems, being quite successful in revenue to high percentage this years. But it doesn't mean next year we don't have new modem coming out. Why do we have new modem coming out? We always provide some benefit to that. So that's point number one, why we still see gross margin have opportunity to go up. Number two, it's to the product mix. Like I say, these years, the [ flex ] products will be P60. And next year, rest assured, we have a new product coming out which is even on a higher [ dimension ] compared to P60. And hopefully, that will bring some positive influence on our gross margin as well. So I think that's the strategy.

[Charlie Chan, analyst]

Okay. Just one last one. The crypto mining ASIC you mentioned in the past quarter, is that still going to take place for this project?

[David Ku, executive]

I think there was a report recently in the newspapers. I think our explanation is, we will not be commenting on the specific product for customer. But on the other hand, I guess, we do have products and we do believe we will ship these products for the crypto minings business.

[Operator]

Ladies and gentlemen, we thank you for all your questions. Now I'm handing it over to Ms. Jessie Wang for closing comments. Ms. Wang, please go ahead.

[Jessie Wang, executive]

Ladies and gentlemen, this concludes MediaTek Inc. second quarter conference call. We would like to thank you for your participation, and you may now disconnect. Thank you.

[Operator]

Yes. We thank you for your participation in today's conference. You may now disconnect. Goodbye.