Okay, Dan, thank you for standing by. Welcome to Ambarella's First Quarter Fiscal Year 2022 Earnings Conference Call. At this time, all participants are in a listen-only mode. After the speaker's presentation, there will be a question and answer session. To ask a question during the session, you will need to press star one on your telephone. Please be advised that today's conference is being recorded. If you require any further assistance, please press star zero. I would now like to hand the conference over to your speaker today, Louis Gerhardy, corporate development and investor relations. Please go ahead.
Thank you, Joelle. Thank you for joining our first quarter fiscal year 2022 financial results conference call for the three months ending April 30th, 2021. With me on the call today is Dr. Fermi Wang, President and CEO, and Casey Eichler, CFO. The primary purpose of today's call is to provide you with information regarding the results for the first quarter of our fiscal year 2022. The discussion today and the responses to your questions will contain forward-looking statements regarding our projected financial results, financial prospects, market growth, and demand for our solutions, among other things. These statements are subject to risks, uncertainties, and assumptions. Should any of these risks or uncertainties materialize, or should our assumptions prove to be incorrect, our actual results could differ materially from these forward-looking statements. We are under no obligation to update these statements.
These risks, uncertainties, and assumptions, as well as other information on potential risk factors that could affect our financial results, are more fully described in the documents we file with the SEC, including the annual report on Form 10-K filed on March 31st, 2021, for fiscal year 2021, ending January 31st, 2021. Access to our first quarter fiscal 2022 results press release, historical results, SEC filings, and a transcript of our prepared remarks, and a replay of today's call can be found on the investor relations portion of our website. Today, we'll begin with a business update from Fermi. I'll review the financial results, then we'll open it up and you can direct questions to Fermi or Casey. With that, I will turn it over to Fermi.
Thank you, Louis. Good afternoon, and thank you for joining us today. The first quarter was another strong quarter, with revenue growing nearly 13% sequentially and 28% on a year-over-year basis. We continued to deliver positive operating leverage, with non-GAAP operating margins expanding to 12% from 8% in the prior quarter and 1% a year ago. Our results are clearly being driven by the beginning of our CV new product cycle and to a much lesser degree by industry-wide cyclical forces. In fact, currently, cyclical forces are actually constraining our results, not exaggerating them. Supply chain challenges remain significant, but our execution is strong, and our guidance assumes the worst of the supply chain challenge from the Texas freeze will be felt in Q2 and gradually improve as we progress through the second half. Capacity is tight, and lead times for certain substrates remain extended.
With solid support from key supply chain partners, including foundry partner Samsung, we usually are not the bottleneck for our customers. With our higher revenue outlook, we remain confident CV will be at least 25% of total revenue this year. Professional security CV revenue, or CV wave one, became material last year, and momentum continues to build, with new design wins and new customers entering production. We expect our home security CV business, or wave two, to become material this year, with several large programs entering mass production now. Automotive, or CV wave three, has clearly commenced. In fact, as of yesterday, May 31st, we have cumulatively shipped more than 450,000 CV SoCs into the automotive market. We expect our automotive business to roughly double this year, and we see tremendous remaining headroom for growth.
Our share exiting this year is expected to be only a few percent of a serviceable market we estimate to be about $2.2 billion in calendar year 2021 and approaching $5 billion in calendar year 2025. I will now update you on our target market progress, beginning with automotive. The automotive market is being transformed by the introduction of electric vehicles that enable sustainable, high-performance transportation. The requirement for advanced active safety features in these vehicles represents a significant new opportunity for our AI vision SoCs due to the need for increasing level of performance in camera-based AI perception. During the quarter, we were excited to see the successful IPO of U.K.-based electric vehicle maker Arrival.
Founded in 2015, Arrival's mission to provide affordable and sustainable urban transportation by producing electric vehicles at a competitive price and by pioneering a new method of manufacturing that challenges traditional economies of scale. I'm pleased to announce that Arrival has selected Ambarella's CVflow AI vision processor for the environmental perception module used to enable L2+ autonomy in Arrival passenger buses and delivery vans. The inclusion of Level 2 + autonomous driving capability provides the driver an array of safety and convenience-related driver assistance, or ADAS, capabilities. Ambarella's AI vision processors were chosen because of their neural network processing performance, stereo vision support, excellent image quality, and extremely low power. We look forward to sharing more information regarding our partnership with Arrival in the near future.
In addition to the efficiency of our CV SoC offer, we have also highlighted the advantage about our open platform approach, which allow OEMs and Tier 1 to create differentiated combo products. Today, I'm pleased to discuss four examples of this from the recent Shanghai Auto Show, namely Great Wall Motors, Momenta, Autocruis, and HASCO. In April, we announced that the leading domestic Chinese SUV automaker, Great Wall Motors, has launched a three-in-one combo system based on our CV25AQ for drive recording, drive monitoring or DMS, and occupancy monitoring or OMS. The system is integral to the new WEY Mocha flagship SUV, which was launched during Auto Shanghai 2021 as the first model from GWM's Coffee Intelligence driving platform.
This CV25AQ-based system can support a variety of simultaneous multi-camera channel combinations for recording and/or DMS and OMS, with the entire system meeting Euro NCAP 2025 standards and playing a key role in GWM's intelligent drive process. Another combo product leveraging our open CV platform is Momenta's AutoRing A4 fleet management solution, combining front ADAS, DMS, and drive recording functions. Based on Ambarella CV22, the product includes front ADAS features such as forward collision warning, pedestrian collision warning, highway monitoring warning, and lane departure warning. While DMS features include face ID, fatigue detection, and distracted driver detection with full HD video recording. Another active safety combo system, this one from automotive Tier 1 Autocruis, is expected to enter mass production this year targeting commercial fleet deployments. The solution combines front ADAS and DMS function on a single CV25 SoC.
Lastly, HASCO, a spin-out of leading OEM, SAIC, demonstrated its adaptive driving beam, or ADB, solution based on Ambarella's CV22AQ SoC. The solution utilized both ADAS and DMS algorithms for intelligent headlight control. As you can see, Ambarella's exhibition at the Shanghai Auto Show in April generated strong interest with more than 50 automotive OEMs and Tier 1s visiting our booth. Ambarella demonstrated a number of designs covering ADAS, electronic mirrors, DMS, and OMS applications, while also demonstrating partnership with many of China's leading third-party automotive software companies. One area of significant interest at the show was our Rebel front ADAS reference design. A turnkey platform based on our CV2FS SoC and the SenseTime software stack.
This solution includes front ADAS features such as pedestrian detection, lane detection, drivable area detection, traffic light, and traffic sign detection, while also supporting millimeter-wave radar and visual perception, providing sensor fusion between the camera and the radar. Our Rebel reference design provides Tier 1 suppliers and software development partners an open platform for differentiated high-performance automotive systems. I will now update you on our continuing progress in the IP security camera markets. During our Q2 fiscal year 2021 earnings call on September 2nd, we stated that in addition to our SoC share gains in professional security cameras outside of China, for the first time, we were also seeing additional opportunities in the professional security market within China. During the last quarter, UNISINSIGHT launched the first of its cameras based on Ambarella solution, including two megapixel and four megapixel models with full color, night vision, and people-counting capabilities.
This is the first camera design in mass production using our new low-cost CV28M and CVflow SoC, which was introduced at the end of last year. During the quarter, KEDACOM introduced its first Ambarella-based design, the IPC695 and IPC445 cameras based on our CV2 and CV22 SoC. The IPC695 features including nine megapixel resolution with HD snapshots and AI-based exposure optimization, while the IPC445 features four megapixel dome design with advanced area intrusion detection and motion detection. Our CVflow SoC are raising the bar in multi-sensor camera designs based on their ability to process multiple high-resolution streams. Concurrently with AI processing. In April, i-PRO, formerly Panasonic Security Camera Business, introduced two new models based on our CV2 SoC, the 8530 and the 8570.
These cameras are equipped with a four sensors per camera capable of independent operation for 360-degree viewing with minimal blind spots and color imaging in just 0.05 lux or extremely low light conditions. In the body-worn security camera category, U.K.-based Reveal became the first to introduce a new camera with Ambarella CVflow SoC to provide a platform for advanced AI-based feature. Based on our CV25 SoC, the new K-series camera includes HD recording, movable lens, and a full-color display. In April, home monitoring market leader Ring, a unit of Amazon, introduced two new models based on Ambarella's CVflow SoCs. The new Video Doorbell Pro 2 model raises the bar for video doorbell designs with 3D motion detection, 1080HD+ video, and the integration of Alexa greeting. Additionally, Ring's new Floodlight Cam Wired Pro also includes 3D motion detection and adverse light video for precise motion alerts.
Lastly, among new customer product introductions, Insta360 introduces a tiny GO 2 action cam based on Ambarella's H22 SoC. The waterproof camera can record 4KP30 video, includes a six times speed hyperlapse mode, image stabilization, and is small enough to be worn on shirts or headbands. From these customer engagements and others, you can see Ambarella's SoCs are enabling customers to add significant value to their products. While efficiency benchmarks like performance per watt and performance per dollar are important elements of customer design win decisions, our open platform and its flexibility also uniquely enable our customer to create optimized and differentiated products and to price them accordingly. Open platform means customer can develop their own software to run our SoC. A flexibility means they can be creative and use the hardware and software resources on our SoC to develop proprietary configurations and feature sets.
Such flexibility is usually not available from our competitors. For example, from automotive market, earlier I discussed projects Great Wall, Momenta, Autocruis, and HASCO that take advantage of the flexibility of our SoCs to create unique combo systems. This powerful solution integrates what used to be two to three discrete camera products with limited functionality into a unified feature-rich solution operating on one Ambarella CV SoC. We are especially excited about our designing with Arrival, where the performance flexibility and low power of our SoC is being harnessed for the next generation of electric vehicles. In conclusion, Ambarella's product portfolio is the strongest in our history. After a very good Q1, despite the supply chain challenges, we are guiding Q2 revenue to be up 48%-54% year-over-year.
Our long-term outlook is fueled by our high level of investment into proprietary technology that is setting the pace of innovation in the visual AI market. Not only is our CV portfolio continuing to extend our reach into new markets, but we believe our visual AI roadmap will enable us to capture more processing value per design win. With that, we are very thankful to have stakeholders, including our dedicated employees, as well as a network of suppliers, customers, and investors who understand the AI vision opportunity and support us as we continue to execute amidst all of the challenges that market has thrown at us in recent years. Once again, thank you.
Thank you, Fermi. I will now review the financial highlights for the first quarter of fiscal year 2022 ending April 30th and provide a financial outlook for our second quarter of fiscal year 2022 ending July 31st. I'll be discussing non-GAAP results and ask that you refer to today's press release for a detailed reconciliation of GAAP to non-GAAP results. For non-GAAP reporting, we have eliminated stock-based compensation expense adjusted for the impact of taxes. Robust demand was capped in the quarter by the supply chain challenges. Revenue of $70.1 million was slightly above the high end of our original guidance. This represents a sequential increase of about 13% from Q4 and a 28% increase from the year ago quarter. Automotive revenue increased about 40% sequentially, security grew more than 20% sequentially, and other product revenue was down more than 25% sequentially.
Non-GAAP gross margin for Q1 was 62.9% compared to 61.4% in the preceding quarter. We incurred some higher costs to manage the supply chain challenges. A relatively stable pricing environment and a more favorable customer mix combined to offset this. Non-GAAP operating expenses for the first quarter were $35.4 million compared to $33.4 million for the previous quarter. Operating expenses increased primarily due to a seasonal increase in payroll taxes and increased headcount. Other income was $593,000, reflecting a continuation of the low interest rate environment. Non-GAAP net income for Q1 was $8.9 million or $0.23 per share compared with non-GAAP net income of $5.1 million or $0.14 per share in the fourth quarter. In the first quarter, the non-GAAP earnings per share were based on 38.1 million diluted shares as compared to 37.6 million in the prior quarter.
Total headcount at the end of the first quarter was 803, with about 81% of employees dedicated to engineering. Approximately 68% of our total headcount is located in Asia. Total accounts receivable at the end of Q1 were $34.5 million, or 44 days of sales outstanding, versus $25 million or 37 days sales outstanding at the end of the prior quarter. The increase was primarily driven by supply constraints, which caused the quarter to be more back-end loaded. Net inventory at the end of the first quarter was $33.1 million, compared to $26.1 million at the end of the previous quarter. Days of inventory increased to 102 days in Q1 from 93 in Q4 for the anticipated demand. In Q1, our operating cash outflow was $4.5 million. Cash and marketable securities were $435.5 million, down from $440.7 million at the end of the fourth quarter.
We had two 10%+ revenue customers in Q1. WT Microelectronics, a fulfillment partner in Taiwan who ships to multiple customers in Asia, was 63% of revenue. Chicony, the Taiwanese ODM manufacturing for multiple customers, came in at 16%. Dahua and Hikvision combined declined sequentially and represented about 10% of our total revenue in Q1. I will now discuss the outlook for the second quarter of fiscal year 2022. During Q2, we expect to continue to experience a variety of supply chain challenges, in particular from the Texas freeze, which disrupted video processor manufacturing at Samsung's Austin, Texas, wafer fab. We expect wafer deliveries from Samsung's Austin fab to gradually recover in the second half of the year, while other industry-wide cyclical forces are likely to constrain supply through the end of the year.
Based on these factors and our best judgment at the current time, we expect total revenue for the second quarter ending July 31st, 2021, to be in the range of $74 million-$77 million. We anticipate both auto and security to increase about 10% sequentially, with other revenue down about 20% sequentially. We estimate Q2 non-GAAP gross margins to be between 61% and 62%, versus 62.9% in the first quarter. We expect to continue to incur higher costs to manage the supply chain challenges, but a stable pricing environment should continue to support gross margins at the mid to high end of our long-term model of 59%-62%. We expect non-GAAP operating expenses in the second quarter to be between $36 million and $37.5 million, due primarily to accelerate hiring and increase chip development costs.
Other income should be modeled $200,000, reflecting lower interest rate on our cash and marketable securities. Q2 non-GAAP tax rate should be modeled in the 3%-6% range. We estimate our diluted share count for Q2 to be approximately 38.3 million shares. Ambarella will be participating in Cowen's TMT Conference and Craig-Hallum's Institutional Investor Conference, both of them tomorrow, June 2nd, Rosenblatt's Age of AI Scaling Conference on June 3rd, Bank of America's Global Technology Conference on June 8th, and Stifel Cross-Sector Insights Conference on June 9th. In addition, in advance of the International Security Conference, also known as ISC West, we'll be hosting virtual demos on June 23rd. Please contact us for more information on those events. With that, I will turn it over to the operator, Joelle, for polling on Q&A.
Thank you. As a reminder, to ask a question, you will need to press star one on your telephone. To withdraw your question, press the pound key. Please stand by while we compile the Q&A roster. Our first question comes from Joe Moore with Morgan Stanley. Your line is now open.
Great. Thank you. Congratulations on the numbers. I wonder if you could talk about the automotive split, since you're saying it's going to more than double this year. What's the split going to be around numbers between OEM and aftermarket, and how much of that do you think will be CV this year?
Right. First of all, I think that the strong growth coming from three areas. First of all, it's from a very strong growth in the OEM drive recorders, which I think in Japan, China, and Korea, we're doing very well. Also, we have a solid growth in the aftermarket, but mainly in the fleet management type of recorder business. That's the first part, is really our existing market. The second growth comes from our view market. We talk about that we have designing with DMS and with ADAS. I think those revenue are ramping up last year and start gradually grow this year, and that definitely inject to the more growth for us in this year. Also, we believe that even Level 2 + and e-Mirror will help us grow the revenue in the very near future. The third thing is really the CV ASP.
We talk about the CV ASP twice higher than the video. This is definitely another factor that we see automotive revenue growth. Although we didn't split out the percentage between CV and non-CV, but we do say that cumulatively, we ship more than 450,000 CV chips from beginning to now. You can see that we continue track shipping quite well in the CV into our automotive market. Hopefully, that answer your question.
All right, great. That's helpful. Thank you. Just for follow-up, in terms of the supply constraints that you guys are seeing in the July quarter, obviously, we've heard some of that from your customers. How much unfulfilled demand do you think there will be, and are you negatively impacting your customer shipments in the month of May, or when does that start to get caught up? Thank you.
Right. The Texas freeze caused problem for the Samsung Tex foundry, and therefore impact us. Like I said, the impact to our Q2 revenue is the worst compared to other quarters, and we believe that the situation will gradually improve in the second half of this year. In terms of the revenue generation, definitely, we impact our customer delivery for sure. We didn't influence on the supplies, in fact, and also we have to go into allocation mode for the video processors, which are produced in the Samsung Texas foundry. The impact is there, although we didn't quantify it, but I think we communicate to all customers who are impacted. We continue to expect the impact will be there for Q3 and Q4, but will be a much lesser degree than Q2.
Great. Thank you.
Thank you. Our next question comes from Vivek Arya with Bank of America. Your line is now open.
Thanks for taking my question. First one, just on gross margin. Good upside in the April quarter, but I think the outlook suggests some reversal of that. Was just hoping for some color around what drove the upside in April and what will drive a different trend in July?
Sure. A large portion of that, again, is our two major customers in China. It's mixed overall, but certainly when their mix goes down, that helps our margin. We did see some of that in the quarter. That would probably be, along with the other margin mixes, that would be the thing that would probably have the highest impact.
Got it. For my follow-up, Fermi, I think you mentioned this very nice number of over 450,000, I believe, CV shipments into automotive. I think the last quarter you mentioned 300,000, so it went up almost 150,000 units sequentially. Could you give us a sense of how much revenue does that represent? Was this exceeding 450,000, was it in line or different than the expectations that you had?
Sorry, first of all, let me clarify the number. Last quarter, we talk about cumulatively 300,000. This week, I talk about cumulatively 450,000. I told you that we ship 150,000 as of today. We are not saying we ship 450,000 this quarter. We talk about 450,000 cumulatively. That's just clarification. In terms of the revenue, I would say that, first of all, the ASP is higher because a lot of it is auto-grade chip, and also the ADAS ASP definitely is much higher than the recorders. Even OEM recorders ASP is higher than the aftermarket. I think overall that the ASP on automotive is pretty healthy for us, and moving forward, we continue to see the same trend. I think this is just showing that the beginning of a CV ramp-up.
We talk about our CV revenue wave, which is wave three, will materialize in 2023. We just try to give you indication that we start seeing this wave moving forward, and we're going to continue to update you about how fast we can ramp up this revenue.
Thank you.
Thank you. Our next question comes from Gary Mobley with Wells Fargo Securities. Your line is now open.
Hey, guys. Let me start out by congratulating you to a strong start to the fiscal year. I appreciate the fact that.
You're able to deliver gross margin upside due to mix and other factors, but I was hoping that maybe you can quantify the supply chain impact, higher substrate cost, higher back-end test and assembly. Could you speak to perhaps the ability of your group to pass along these price increases or renegotiate some of these price increases for your customers?
Right. We did see a price increase on the substrates and to a certain extent on the testing packaging side. For those price increase, we have not reflected in our ASP yet. We believe that at this point that we will not change our ASP because of that. We might reconsider this position if our wafer price got changed. Until then, we feel that we're comfortable with the decision. Our only price change we did through the last few months was that, because of the allocation, some of the customer are willing to pay for super hot lot to get a bigger allocation, and we help them to get those super hot lot by paying more money. We ask our customer to share the increase on those super hot lot side. That's the only ASP we grow, reflect to our customer.
Appreciate that. As my follow-up question, I wanted to ask about your product roadmap direction. If I'm not mistaken, you, I think, envision maybe rolling in some sensor fusion domain controller technology, and I was hoping that you can give us an update on where this initiative stands.
As you can see, we have addressed. If you look at the automotive CV market, in fact, we have a solution for all of them except domain controller. It's our ambition and we need to have a domain controller type of solution to not only just for sensor fusion, but also to provide the higher performance at much lower power consumption than our competitor can offer in the near future. We are developing that, and hopefully we can give you update soon.
Appreciate it. Thanks, guys.
Thank you. Our next question comes from Kevin Cassidy with Rosenblatt. Your line is now open.
Thanks. Congratulations on the strong quarter. You mentioned the wave two happening this year with some large programs, and if that's home security cameras. Can you say that are some of those programs launched and, I guess, what percentage would you say you're into? Is it going to be a stronger October quarter compared to this July quarter?
I think in my script just now, I mentioned that Ring announced two products with our CVflow SoCs, and they are basically replacing our existing Ring products with video-only solutions. Yes, I think that we continue to see strong momentum CVflow designs, and also we expect to see other consumer IP security camera customer introducing CVflow-based solution in the near future and going to production this year.
Okay. Are there still any human vision designs or are all the new designs CV designs?
We still see several human vision design wins. I would say that it's more towards the lower end. All the middle and high-end solutions targeting CV these days. We still see a mix of CV and non-CV products based on our customers, but we continue to focus more on the CV side because that's where our strength is and where we have differentiation.
Okay, great. Thank you.
Thank you. Our next question comes from Tore Svanberg with Stifel. Your line is now open.
Yes, thank you, and congratulations on the results. First question is back to auto and some of the subsegments. If I can maybe just split between recorder and non-recorder, is it safe to say that this year recorder would be about 90% with a lot of the new applications being the other 10%, roughly?
I think that 90% too high for recorders. I don't have a number with me, but gut feeling tells me that our other business will be more than 10%.
Great. Thank you. As my follow-up, I know in the past you've talked about the amount of cameras going into car is about 1.5 in last year. Do you have any estimate this year based on conversations you're having with your customers, how that number could potentially turn out?
Yeah, Tore, the 1.3 is like a third-party market research firm estimate across all new vehicles produced in a year. That number, the expectation is that it's going to continue to increase as more and more cameras are used in the car, not just human viewing cameras, but in particular, more and more of the sensing cameras that require the CVflow chips for things such as driver monitoring or front ADAS or surround view or all kinds of other applications. We see that number, third-party firms continue to see that number increasing over the next several years.
Great. Thank you, and congrats again.
Thank you. Our next question comes from Andrew Buscaglia with Berenberg. Your line is now open.
Hi, guys. Did you guys give a number around exactly what CV represented as a percentage of sales in the quarter? Given strength in automotive, do you still foresee 25% of your sales being computer vision sales this year, or has that expectation increased?
First of all, we still expect that our CV revenue will be more than 25% of total revenue this year. We think that we have a very good chance to deliver on that. With automotive compute increase, that definitely will help us to drive more CV growth. As you can see that the majority of CV revenue still comes from security, particularly professional security camera. I think that will be a continued driver for CV until next year when consumer IP can jump up and also more and more automotive CV revenue coming in. I would say next year we can see continual growth on our CV revenue.
Okay. This is your fourth quarter in a row with at the upper band of that gross margin range, and you're guided to kind of upper band at the midpoint. Really, your long-term guidance would imply a kind of a sizable step down in the back half if you still want to fit in that 59%-62% range. I guess, what do you need to see or what would provide you some confidence to maybe move the low end of that up at some point? It just seems like you keep outperforming your own. I know you have some mix issues, but if you can talk to me about that.
Certainly part of it is mix, and a part of it is the uncertainty that Fermi talked about for the second half of the year, depending on how that plays out. There's pricing issues and other issues that could create that to move. Again, we've been at the high end of the range as you commented, and we'll take that. We felt that with everything that's going on right now, we needed to be balanced in the way we're looking out over the second half of the year, or at least certainly the quarter to make sure that we keep our range to where it is today. If we see something different longer term, well, then we'll start to think about that.
All right. Thank you, guys.
Thank you. Our next question comes from Matt Ramsay with Cowen. Your line is now open.
Thank you very much, guys. Good afternoon. Fermi, I was particularly interested in the announcement that you made with Arrival, the CV2 FS win. It sounds like you might be going to give us some color as you're able going forward, but if you could give a little bit more color on the background there, and what were the folks that you were competing against and the particular features that allowed you to take that big design win down? Thank you.
I think we're competing with everybody out there, and I truly believe Arrival beat all of the people who has a similar solution out there. I think CV2FS was chosen for several reasons. I think the biggest reason is our performance per watt, and also it's an open platform that really enabled them to do a lot of things on their own. I think those two things combined help us to get the design wins.
Got it. Thanks. I appreciate that. My follow-up question is for Casey. Obviously, the primary drivers of Ambarella going forward are going to be in auto and security, but I noticed that in the printed results, the other category was down, I don't know, on the order of 20%, I think you mentioned, and then it looks like it's down 20% again. I'd be interested if there are things changing that you might call out in some of those other smaller markets from a supply-demand perspective, or is this supply tightness at Samsung and you guys allocating wafers to your more strategic business, or is it a combination of both? Thanks.
Yeah, it really is a combination of both. When we talked about other a couple of years ago, we expected that to start declining and continue to decline over three to five years. We did have some upside in the past year that we took and we're happy about. As we look forward, because we're moving everything towards the CV product line in a lot of those markets, the CV really doesn't have the advantages it does in some of the markets we've talked about. We'd anticipate that to continue to decline over time. A lot of that historically in the past few quarters have been just a few customers that really have driven that. DJI, we've talked about in the past and a few others.
As we go forward, if we see some opportunities or markets that CV is well suited for, well, then I think you might be able to see some opportunity there. But meanwhile, we kind of thought about it as a declining market, and if we get upside, we've taken it, but we aren't really changing our viewpoint on that.
Got it. Thanks, guys. Appreciate it.
You bet.
Thank you. Our next question comes from Ross Seymore with Deutsche Bank. Your line is now open.
Hi, guys. Congrats on the strong results. Just had a couple questions, but the first one is on the inventory and the supply side of things. How did you guys have your inventory increase so significantly? Not that the days are out of whack with anything historically, but I was a little surprised that you have those constraints on one hand that you talked about with revenue in the second quarter being the worst, but your inventory still went up, I think, about 25%, 30% sequentially. Is that just different parts where you're not supply constrained outside of the Austin fab or specific end markets likely to be more impacted by the Austin freeze? Any color there would be helpful.
Well, first of all, I think that we have very good visibility for our customers, and the demand is strong. Basically, the inventory number you see is based on, first of all, the strong demand. Second, I think it's become very clear that you need to give a bigger lead time. Because the lead time increased a lot, we decide to increase our inventory level so that we can protect ourselves and protect our customers. The combination of these two really drive our inventory. At the same time, I have to say, I think we are confident those inventories are not going to be dead inventory for us because we know the run rate of the customer. We have multiple customers in the different product lines, and we are confident that we're building an inventory that we will be able to sell.
The most important thing is considering the current environment, we really need to make sure that we have enough product for our customers if the lead time of those products continues to be a problem for us.
Thanks for that, Fermi. I guess for my follow-up question, kind of a two-part one here, and it's a little bit following on a prior question that was asked on your other business. I'm really getting at what seasonality can look like in the back half of the year if you put together the supply loosening up on the positive side and all the design wins you have, but the potential negative side being the fact that the consumer side is just not as big as your business. The two questions would be, how would you think about second-half seasonality, and any sort of ballpark as a percentage of sales where that consumer business is in the first half of the year? It seems like it must be down to, I don't know, right around a 10% number.
Any kind of right-sizing there for our models would be helpful.
Yeah, I think when we look at the second half of the year, as we've talked about, there's a lot of uncertainty in the second half of the year. When we look out, I think we feel opportunity to continue to supply to our customers barring any change in some of the dynamics that we're seeing today. I think we don't guide beyond one quarter, but I think that we've seen the worst of it, as Fermi talked about earlier, in a lot of the areas, and we should be able to meet demand as it comes through in the second half of the year. We're also looking into some of the newer markets that we talked about. As we said, we've got three phases.
By the end of this year into next year, we're going to have all phases kind of starting to really kick in, and that's what's really exciting to us.
Great. Do you need sort of right sizing on how big that consumer is as a percentage of sales? Was it kind of around 10%-ish in the first quarter, and then you're implying your guidance in the second?
You mean the consumer business that was down?
Just trying to see. You gave percentage sequential changes for the quarter and the guide, just want to make sure the absolutes are aligned as well.
Yeah, I think you're going to see the numbers go down in that market as well. You're going to see the other two go up, and you're going to see the consumer business go down as we talked about.
Okay, one final housekeeping one. Fermi Wang, you had talked about Dahua and Hikvision being, I think, low to mid-teens two quarters ago. You said it dropped to about 10% in the most recent quarter. Can you just talk a little bit about did the dollars drop as well as the percentages? More importantly, what's the update on Dahua rolling out the CV design wins that you have had? Is that a customer that you expect to be a tailwind going forward, or is there some change in the narrative there?
Right. For Hikvision and Dahua, both of them are using our video products, and we talk about they build up inventory a few quarters ago. They just, I think, digest the inventory and start reordering for some of the parts, and they continue to reorder, but much lower levels. I think our video product sales to those two companies will continue to shrink over time. For Dahua, we already ramp up our CV revenue with them, but it's kind of limited by supply as well as by the substrate situation. I think as soon as we transition out of those problems, we think that Dahua CV revenue will be a growth for us. Overall, I still think that combined, the dollar sign probably is similar to our last quarter.
Great. Thank you.
Thank you. Our next question comes from Suji Des ilva with Roth. Your line is now open.
Hi, Fermi, Casey, Louis. Congratulations on the progress here. Perhaps first for Casey on the operating leverage here, you got the nice operating margin bump up to 12% here. Can you talk about the expectations for OpEx and what you think you'll be able to do as you continue to grow the revenue? Will you need to invest more here, or can you keep that relatively stable?
I think we'll continue to invest, not only in people but also in opportunity. I think we'll see that. Having said that, we're kind of focused on getting to 20% right now. As you know, the business can, in the past, has done 20%-25%, and we think that's still attainable as well. Right now, from where we are, we're focused on getting to those type of margins. To your point, we've kind of grown pretty well over the last few quarters, and we anticipate we should be able to do that into the next few as well.
Okay, great. A question for Fermi. Perhaps a lot of press on autonomous attempts here and the notion of camera plus radar, perhaps shifting to camera plus LiDAR, and just what camera can and can't do. Maybe you could talk about the implications for Ambarella's product designs and automotive footprint and whether that is impactful to you, whether you're kind of we're talking to all that.
Right. First of all, we continue to believe that camera is the most important sensor in any kind of autonomous driving market. Also, in front of us, the requirement for the camera technology will continue to grow. We're going to get more camera per car. Every camera will have a higher resolution, higher frame rate. You need stereo processing. You need a much better CV performance to offer more and more different type of detections, and you need to have a sense of fusion. With that, I really think the camera, we're going to continue to have a good roadmap for our camera products for autonomous driving. Having said that, I also believe that redundancy is very important for those market. We believe that radar particularly is a good partner for any camera solutions.
I think the reason I didn't say LiDAR is because still on the price side, LiDAR is expensive and radar is more accessible, radar is not optical sensor, that is really a good technology complement to our camera technologies.
Okay. Thanks, guys.
You bet.
Thank you. Our next question comes from Quinn Bolton with Needham & Company. Your line is now open.
Hey, guys. All right . My congratulations. Just wanted to come back to the Samsung Austin impact on the business. Did you say Fermi that that was just for the video processor side of the business, or are there any CV products manufactured out of that facility?
Right. It's only a video processor. All our CV processor are built in Korea factories.
Got it. Thanks for that clarification. I guess with that mix shift, if you're constrained or most constrained on the video processor side, wouldn't that be a tailwind to margins over the next couple of quarters?
No. Margins basically are the same between CV and the vision-based products. While obviously, we talked about the fact that it doubles the top line, it doesn't really change the margin at all.
Got it. Okay. Lastly, just with the video processors being constrained, do you think any of that demand that you might not meet this quarter or next is perishable, or would you expect that demand just to sort of roll into future quarters and you'll meet it when you can get the wafers or substrates?
I assume it's perishable. If you look at those video products, some of them go to professional security cameras, some go to consumer security camera, some go to other consumer market like DJI. For the consumer security camera and the consumer camera, those demand is perishable. For professional security camera, it may be there, but still, there are some requirements on different timing. For example, usually not selling season, the people at the end of the year try to digest their budget, and usually buy a lot in Q4. If we miss those points, those demand might be also perishable.
Understood. Thanks for that clarification.
Thank you. Our next question comes from Tristan Gerra with Baird. Your line is now open.
Hi, good afternoon. Just a quick follow-up on the gross margin outlook. You've mentioned some of the catalysts that might have an impact on gross margin directionally for the second half of the year. As your supply improves, is it fair to assume that your surveillance camera customers in China are going to improve as a percent of total mix, and that's going to be a gross margin component to consider for the second half? Also, if you could give us an update on the impact of 5-nm development cost on gross margin for the next few quarters.
Yeah. Well, like I said, we're not changing our standard guidance that we've had from the margin perspective. We guide quarter by quarter, so we've given that indication. When we get out further, I think that you can see some different mix. As we talked about earlier, again, Japan in particular has a mix issue that can impact us from quarter to quarter, and as we look out into the next couple of quarters, we're going to have to see how that comes out for us. There's several different mix elements, but that's certainly the biggest one that we've seen historically over the last maybe four to six quarters.
Okay. Just as a quick follow-up, any changes that you see in terms of the competitive environment in China, in surveillance, notably at the low end? I am assuming that would not impact you, but do you expect any changes in the pricing competitive environment medium-term?
Right. In terms of the competitive situation in China, I think HiSilicon is meant to be things that we need to watch out, because first of all, they still have tons of inventory sitting either in a customer or within HiSilicon. We still seeing them shipping the products. Other than HiSilicon, we haven't seen a competitor that in the mainstream or high-end product line. There are many, many, many low-end Chinese to Taiwanese competitors competing in the low-end video and even some low-end CV side. I think that price pressure is always there on the lower end side, but on the middle and the high-end side, I think we are comfortable with our current ASP guidance.
Okay. That begs just one last quick follow-up, which is, when do you think HiSilicon inventories might normalize, and would that translate into an acceleration in demand for your product?
We don't know. However, our customer gives us indication they can easily survive for another 12 months without any new products in. I think they're sitting on tons of inventory right now.
Great. Thank you very much.
Thank you. Our next question comes from Richard Shannon with Craig-Hallum.
Well, thanks guys for taking my question. Fermi, maybe kind of an industry question here. Professional security as it relates to CV. How are you seeing the adoption curves here relative to what you saw in the past transitions from analog to digital? Is it faster, slower? What kind of indications are you seeing in market pricing here that would indicate either of those inflections?
Right. First of all, when we went through this analog to digital IP security transition 10 years ago, I remember that first two years it was slow, then suddenly it transitioned. I would say that we are not in a hockey stick yet. We haven't seen that, because the majority of our professional security camera unit numbers shipment still in a video-only solution. With that, I would say that we're still at the early stage of the transition from video to CV. I think the biggest reason that the trigger of that transition, if the hockey stick need to happen, I think that customer need to drive their price, CV camera product price, to maybe only a minimum premium over the current video product. That will trigger the transition, and I don't think we're there yet.
Okay. That's a helpful perspective, Fermi. Thanks for that. My last question here is in the automotive side, again, related to CV, applications like DMS and ADAS, et cetera. Not looking so much at the higher-end ones like L4 and L5, but more the immediate-term ones. Any way you have a sense of, and can you characterize for us, what kind of share and design wins do you think you have in that space right now or have visibility into?
We don't have a number for you. However, I can say that I think if you read through our design win activity in the last few quarter, every quarter we publish those design wins. You can see that we have a consistent, maybe even increasing number of design wins on the DMS and OMS and the ADAS now. Obviously our market share is increasing on those near-term market, but I just don't have a number because most of them are still pretty new. I would still say that in the ADAS market, Mobileye is still dominant, and we still will just start gaining some design wins in there. I hope that momentum will continue to carry us moving forward. I would say on ADAS, we're still very small.
From the OMS, DMS side, because it's brand new market, I think we have a chance to get a bigger market share there.
Okay. Fermi, any way you'd want to characterize what your goals might be in OMS and DMS? Obviously, it's difficult competing strictly against Mobileye, but could you get to 20%-25% of that market over time?
In fact, our goal, the thing we really want to push is not just capturing market share, DMS, OMS. We are trying to push the concept that our customer Tier 1 OEM, which have ADAS, have recorders as DMS and OMS. They need one unified platform to do all of that. If you look at ADAS, it is really a front-facing camera to do mobile ADAS, and also front-facing camera to do a recorder today. There is no reason these two product cannot be the same. At the same time, the DMS is really at the same location you have of ADAS camera. You are facing inside and looking at the driver. We like to argue the driver and the ADAS need to be synced in the video so that when something happens, you know exactly what the driver is doing. That has to be synced.
I really think the combo product makes sense, and we really think if that scenario become a market requirement, that will definitely help us to gain market share, because I don't see any other competitor solution can do those kind of combo product today.
Excellent perspective. I think I'll follow up online with that one, but thank you very much, Fermi. It's helpful.
Thank you.
Thank you. Our next question comes from Tore Svanberg with Stifel. Your line is now open.
Yeah, just a quick follow-up and maybe related to the last topic there. Are you still on track to sample in CV5 this quarter, and is everything sort of on track with Samsung's 5-nm process there?
Yes. The answer is yes and yes, and we are on track for both. We are working on CV52 in our lab right now. We believe we're getting ready to sample it. Also, we are working on, of course, the CV5. CV5 is our first 5-nm chip with Samsung, and we continue to work with Samsung on the CV5 process improvement as well as the production plan. I continue to believe we will start generating CV5 revenue sometime next year.
That's great. Thank you.
Thank you. Our next question comes from Martin Yang with Oppenheimer. Your line is now open.
Hi. Thanks for taking my question. Just one from me. I'm curious if current supply constraints may affect or change how HiSilicon can think about its adoption for CV. Is there anything that may change their view on maybe getting to CV sooner? Do you have a view there?
Well, you're talking about Hikvision.
Okay.
We talk to Hikvision all the time. I think that Hikvision is really following the direction that try not to use U.S. component as much as they can. Actually, this is probably one company we see, they're following policy very tightly. I think that's why they build up so much HiSilicon inventory, and I believe they are trying to find out whether they have options with other Chinese supplier. I think that's the current situation. Like I said before, we cannot force people to look at U.S. component. The only thing we can help our customer is to provide a solution that is very hard to find a replacement in China, and that's how we can get continue to have a multi sharing [audio distortion].
Thank you.
Thank you. Our next question comes from Derek Soderberg with Colliers Securities. Your line is now open.
Hi, guys. Just one question from me as well. Fermi, I'm wondering, in your view, which design wins you announced are the most meaningful for the company? If you could maybe just call out the top two or three wins and provide any additional detail on those wins, I think that'd be helpful. Thanks.
You're talking about design win that we just mentioned. I think Arrival is obviously the most important one for me. Not only that's an automotive design win, but they are important for different reason. One, this is a Level 2+ and using our CVflow engine. More importantly, this is for the first CV2FS chip, that design win we announced. That just show you that our CV2FS functional safety chip is in a very good shape, and our customer is already adopting it and testing it. I really think that just from that point of view, I'm excited, but not to mention that Arrival is a great European company, and they attract a lot of attention from different places. I think that's definitely important for us. The other one that equally excited about is this Great Wall announcement that we had just a month ago.
Great Wall Motor is really a very well-known Chinese OEMs, and they adopt our solution. It's actually a combo solution, by the way. It's really a drive recorder with DMS, with OMS all in one, and they are the first one to really push this concept, and hopefully with this, and we show our capability and performance with GWM, and other people picked it up, said this is right way to go, and hopefully that will give us more design wins in the future. For these reasons, so these two are probably the project I'm most excited about.
Perfect. Thanks.
Thank you. I'm not showing any further questions at this time. I would now like to turn the call back over to Dr. Fermi Wang for closing remarks.
First of all, thank you very much for joining us today. I'm looking forward to see you sometime next time. Thank you. Bye.
This concludes today's conference call. Thank you for participating. You may now disconnect.