Artificer60
Well-known member
Maybe I'm missing something here, but I feel like this is the type of calculation that lead Otellini to pass on making chips for the Apple iPhone.
Intel: New Products must generate 50% gross margin to get the green light
- Thread starter Xebec
- Start date
siliconbruh999
Well-known member
Exactly no risk no reward
hist78
Well-known member
If a 50% gross profit margin is so easy to achieve, we should have seen many semiconductor companies doing that. I think it's a tough and tricky requirement.
I started thinking about TSMC's journey into the advanced packaging services.
"It's interesting how important TSMC's advanced packaging solutions, like CoWoS, have become for companies like TSMC, Nvidia, Apple, AMD, and many others, driving their products, revenue, and profit. It wasn’t like this when TSMC first introduced its advanced packaging solutions about 14-15 years ago. In fact, Nvidia was one of the companies that rejected the idea back in those early days.
TSMC's first win in advanced packaging came from Xilinx, which placed an order for just 50 wafers per month. Yes, a whopping 50 wafers per month!
TSMC's former CTO, Shang-Yi Chiang, recalled that he became the subject of laughter at TSMC for investing about $100 million and 400 engineers into the R&D for advanced packaging—only to secure an order of 50 wafers per month."
Source: https://semiwiki.com/forum/threads/...says-morgan-stanley-–-report.21080/post-75439
Last edited:
I think that is the target. Then, IFS can provide the product team with same leeway.
MKWVentures
Moderator
how is 50% GM limiting?
Products always plan for this, they adjust the price for that in planning stage. then the reality of cost and pricing comes in and the CEO says "due to unforeseen ramp costs...."
AMD has 50% margin at TSMC, nvidia has 70% margin at TSMC, broadcom has 50% GM. Intel used to have way above 50% GM internal. Somehow, Intel has managed to lower its margins....
and TSMC has 50% GM itself.
If any Intel product does not at least PLAN for 50% margin, it should be killed. How can Intel fail to match AMD, broadcom, etc?
Products always plan for this, they adjust the price for that in planning stage. then the reality of cost and pricing comes in and the CEO says "due to unforeseen ramp costs...."
AMD has 50% margin at TSMC, nvidia has 70% margin at TSMC, broadcom has 50% GM. Intel used to have way above 50% GM internal. Somehow, Intel has managed to lower its margins....
and TSMC has 50% GM itself.
If any Intel product does not at least PLAN for 50% margin, it should be killed. How can Intel fail to match AMD, broadcom, etc?
hist78
Well-known member
Are we talking about a 50% gross profit margin for each individual product, or the combined gross profit from all products divided by the total company revenue?
From the news report, it seems Intel is expecting each new product to meet this 50% gross profit margin requirement. I don't think that's reasonable or practical.
50% gross margin at the product level is quite low for Intel historically, and actually low for high volume chip product lines from established merchant chip vendors.
MKWVentures
Moderator
50% GM on each product during product approval phases. that is very reasonable.... This really isnt new. the CEO can approve variations.
the main issue is manufacturing cost. if intel can be competitive in manufacturing this is easy. AMD, Broadcom, Nvidia, Apple do this and they outsource everything.
I have seen Intel approve products in the past that had negative NPV and negative margins. then killed them late in the game .... talk about waste.
hist78
Well-known member
It sounds great on paper to have the CEO to consider any exceptions, but don't you think this guideline could kill many potentially profitable projects, especially those that may not generate huge profits right away? In a typical corporate environment, most people will simply follow the policy and have little motivation to challenge top executives for the sake of a single project. It's just not worth the risk.
Intel has always done a lot of packaging in-house, yet it's TSMC’s advanced packaging that’s currently making tons of money. Apple once asked Intel to manufacture processors for iPhones, but Intel turned down what could have been a once-in-a-century opportunity. OpenAI also approached Intel to become a major investor and processor supplier, yet again, Intel rejected the offer. Clearly, Intel didn’t see enough profit potential in these opportunities to get involved.
I suspect Intel still holds onto the old IDM mentality, believing it should be the dominant party that sets both price and profit. Unfortunately, in today’s competitive environment, I don’t think Intel can afford to be that selective anymore. Survival may depend on adaptability and Intel's first priority right now is to survive.
siliconbruh999
Well-known member
Maybe but gross margin doesn't count R&D and other expenses it only counts the cost of production
You forgot there was no one competent at the CEO position .
There is nothing wrong with IDM if you operate both arm correctly tbf memory makers are IDM/TI is an IDM as well.
MKWVentures
Moderator
Intel needs to kill some potentially profitable projects in exchange for more profitable projects where they can actually execute IMO.
The Product reviews have lifetime NPV and GM and all the variables and scenarios
I think people need to come to the reality: Intel cannot afford all their projects until they fix the financial problems. It is not currently sustainable. Expanding spending on tons of Fabs, sellling off interest in leading fabs, and having revenue shrink (during a massive market growth) is a self inflicted injury that Tan now needs to fix (just an opinion)
Intel has had many projects that should have been canceled and many equally that if properly executed, managed, and resourced that could have expanded their portfolio.
Sadly comes down to management and culture.
All I see now is same old same old with a lot of culture and customer focus talk and layoffs, reality leaders must drive and show action to change the culture, is it happening ?
Sadly comes down to management and culture.
All I see now is same old same old with a lot of culture and customer focus talk and layoffs, reality leaders must drive and show action to change the culture, is it happening ?
I think it really all depends on how that 50% is calculated... For tax reasons, shareholder value, or corporate ego - it could be either really easy or really hard to hit 50% Gross Margin for Intel Products, depending on how Foundry costs enter the discussion. Is the 50% GM for Products only, or for the whole of Intel?
Also, Is there an allowance for a year or two of underperformance to gain market share when entering a new market?
A personal analogy...
My father/family owned (and some family still owns) a wholesale and a retail business. 50% of the wholesale's business is to the retail stores the family owns (and 50% is to external customers). Sometimes the family had the wholesale sell to retail at a small loss (i.e. -2%*) , and other times the family decided the wholesale should have a margin equal to a typical wholesale business (i.e. 15%*). This flipping around made it hard to "at a glance" determine the financial health and performance of individual stores as the "real GM" changed over time depending on the formula between wholesale and retail.
Take this a step further to a large corporation like Intel, with stacks of bean counters, er.. financial professionals, on both businesses -- and the tax complexities that hit large global enterprises, and a single statement of "all products must have 50% GM" is not super meaningful.
..
*I don't remember at all what the actual percentages were, these are just examples.
I doubt your family business was a public company audited by an accounting firm like Ernst & Young.
For public companies (those selling shares to the public), financial games like you describe would not be sanctioned by the auditor.There is a difference between product-level gross margins, which are used for internal decision-making, and the corporate gross margin reported in financial reports, like a 10-Q. Internal decision-making can use any sort of calculations management decides to use. For internally fab'd chips cost recognition rules were up to management, but for corporate gross margin calculations, there are specific accounting rules.
It's a dual-edged sword. It's also the same calculation that kept them out of the iPhone and out of the discrete GPU business when it would have mattered most to entering leading edge silicon markets.
I doubt your family business was a public company audited by an accounting firm like Ernst & Young.
There is a difference between product-level gross margins, which are used for internal decision-making, and the corporate gross margin reported in financial reports, like a 10-Q. Internal decision-making can use any sort of calculations management decides to use. For internally fab'd chips cost recognition rules were up to management, but for corporate gross margin calculations, there are specific accounting rules.
Enron was audited by public financial firm Arthur Anderson
There still is going to be a lot of room for 50% being a 'pessimistic' or 'optimistic' number, not just based on the sales pitch internally, but also on how they choose to 'cost out' Intel Foundry vs. External foundries.. Unless we know that what the assumptions are for the foundry costs, 50% GM is .. +/- a lot.
MKWVentures
Moderator
this is already decided for intel today. its 50% gross margin for the product side. foundry sells to them at market price. taxes, shareholder value, corporate ego have nothing to do with it.
they claim that foundry and product are separate and independent. If they are not, given what they have officially said, that is problematic.
50% GM using a foundry is not hard for good companies
Thanks! Ok so the product team can choose whichever price is better (internal or external foundry) to hit their 50%?
That still gives an opportunity for the product team to push for a few % savings on their end that might be worse for the overall Intel umbrella. (I.e. the reduction in business for Intel foundry is a larger hit overall than the pennies saved per product).
I've dealt with large corporations that are actually multiple companies under one banner long enough to know there's always a way for corporate egos to make a decision for one division that isn't the best for the whole
. (You guys are way more "glass half full" on this than me
).The iPhone decision was most about the necessary CAPEX for the fabs, magnified by Job's low unit price demand. Even in retrospect, fabricating A-series chips never looked like a winning plan for Intel. Intel had the wrong cost structure for anything but high-margin (to Intel) chips.
The lack of a discrete GPU strategy was not a gross margin issue. It was a "we don't need them because we're going to win with integrated graphics" for clients, and datacenter GPUs were a niche, and small one at that, with annoying SIMD systolic array architectures that made finding a killer app daunting until LLM training came along. I remember when GPU-accelerated database management software was thought to be next big datacenter thing. That turned out to be a small niche. Supercomputers, which used to be the realm of systolic arrays, went to scale-out x86, even rolling over Intel's own i960 in the process. If it wasn't an x86 CPU, it wasn't a good plan. By the time datacenter GPUs got hot in the market, Intel was (IMO) too far behind to develop the product maturity to catch up. And Xe HPC is not an easy architecture to evolve. Have you ever looked at it closely? This is one complicated beast of a product line.
Intel® Xe GPU Architecture
Programming oneAPI projects to maximize hardware abilities.
www.intel.com