Discussion Intel - the cost of BACKPORTING

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Kocicak

Senior member
Jan 17, 2019
982
973
136
Seeing the preliminary results in the Anandtech 11700K review, the performance of this CPU compared to the previous CPU overall seems to be just marginally better, in some cases even worse than the previous generation product.

What is the economic sense and impact on the company of the decision to backport this new CPU to the 14 nm technology? How much money does this backport actually cost? Was this cost really worth it, when the only result is that you will have some "fill in product" for a few months before the Alder lake comes?

If it does not make sense financialy, then why did they do it? They must have known well that the end result of this will be.
 
  • Like
Reactions: krumme

dmens

Platinum Member
Mar 18, 2005
2,271
917
136
Decoupling was EXACTLY done so you could backport. Bringing it to a next gen process is much easier. It was very clear from the words used in the interview that they wanted to prevent a future repetition of stagnation in architecture when f.i. 7nm also turns out problematic.

Nah man. Given enough die area and the willingness to take a hit on power and frequency, backporting can start at any time. Architecture, RTL and schematics are process agnostic. The issue with the Intel design has always been its tools/methodology being in-house and only working with the Intel process.
 

TheELF

Diamond Member
Dec 22, 2012
3,973
730
126
Well, 5 years ago is long time ago and not very relevant for today. In the last couple of years AMD is outselling Intel CPUs in PC DIY market significantly.
This is the yearly revenue of intel from 2015 where they had no competition at all to last year where the aging skylake was competing against ryzen 5000.
DIY market is either extremely small and irrelevant or intel is selling a good share towards DIY as well.

2020$77,867
2019$71,965
2018$70,848
2017$62,761
2016$59,387
2015$55,355
 

TheELF

Diamond Member
Dec 22, 2012
3,973
730
126
Intel is in the Cash Cow years if they do not change. They will make good money on what is a legacy based product but long term will lose out in the future.
Yes but for this to happen there has to be somebody able to step in and take over, the rate AMD is going they might be able to do it in like 20 years or so, AMD is up to 1bil income while intel is still at almost 21 bil.
 

Kocicak

Senior member
Jan 17, 2019
982
973
136
I have no idea how total revenue of the whole company is relevant in this topic, perhaps you are trying to say that Intel can afford to cover the costs of this "Rocket adventure"?
 

dullard

Elite Member
May 21, 2001
25,059
3,410
126
I have no idea how total revenue of the whole company is relevant in this topic, perhaps you are trying to say that Intel can afford to cover the costs of this "Rocket adventure"?
You answered your own question.

I honestly don't know the cost of this backport. I'll assume jpiniero is correct in post #2 that it is in the 10s of millions. But on the scale of Intel, that reports 72000 million in revenue in a year, and profit is rounded to the nearest 100 million, 10s of millions is less than a rounding error.


I think Rocket Lake makes much more sense in an OEM market for middle-powered computers. The iGPU is just enough to be worthy of cutting out the video cards which not only save money for the OEM but video cards are hard to come by right now. In that sense, the money might very well have been worth it for Intel. Is Rocket Lake a great chip? No. Is it just enough for Intel to fill in a void until 10 nm desktop is a reality? Probably.
 
  • Like
Reactions: Tlh97 and Hulk

Hulk

Diamond Member
Oct 9, 1999
4,223
2,012
136
You answered your own question.

I honestly don't know the cost of this backport. I'll assume jpiniero is correct in post #2 that it is in the 10s of millions. But on the scale of Intel, that reports 72000 million in revenue in a year, and profit is rounded to the nearest 100 million, 10s of millions is less than a rounding error.


I think Rocket Lake makes much more sense in an OEM market for middle-powered computers. The iGPU is just enough to be worthy of cutting out the video cards which not only save money for the OEM but video cards are hard to come by right now. In that sense, the money might very well have been worth it for Intel. Is Rocket Lake a great chip? No. Is it just enough for Intel to fill in a void until 10 nm desktop is a reality? Probably.

Good point regarding the inclusion of the GPU for the "light" gaming crowd. I wonder if Intel could have squeezed in 64 or 96 EU's in the GPU and made decent gaming a reality? As you wrote, considering the GPU market that might have made sense for quite a few people.
 

TheELF

Diamond Member
Dec 22, 2012
3,973
730
126
Good point regarding the inclusion of the GPU for the "light" gaming crowd. I wonder if Intel could have squeezed in 64 or 96 EU's in the GPU and made decent gaming a reality? As you wrote, considering the GPU market that might have made sense for quite a few people.
Intel is still serious about releasing a whole range of discreet GPUs and making the integrated too strong would seriously cut into their lower range cards.
 

TheELF

Diamond Member
Dec 22, 2012
3,973
730
126
I have no idea how total revenue of the whole company is relevant in this topic, perhaps you are trying to say that Intel can afford to cover the costs of this "Rocket adventure"?
How is AMD outselling intel relevant if it doesn't influence the sales of intel at all?!
 

Failnaught

Member
Aug 4, 2008
26
25
91
So what is the reason for the low performance on a technical level? Is it big physical dies don't perform well (but Xeon is big and does fine)? Or it was half-baked and not optimized to the 14nm process in some way?
 
  • Like
Reactions: ryan20fun

TheELF

Diamond Member
Dec 22, 2012
3,973
730
126
So what is the reason for the low performance on a technical level? Is it big physical dies don't perform well (but Xeon is big and does fine)? Or it was half-baked and not optimized to the 14nm process in some way?
Intel makes desktop CPUs, those are made to do well with desktop workloads but include an iGPU that can handle some amount of workstation workloads as well, Xeon as well as Zen are workstation CPUs, they are made to do well with workstation workloads so they don't have any iGPU to help them with workstation workloads.
 

Kocicak

Senior member
Jan 17, 2019
982
973
136
How is AMD outselling intel relevant if it doesn't influence the sales of intel at all?!

There is a lot of evidence, that sales of Intel CPUs for DIY market are down significantly. Have you noticed?

Intel makes desktop CPUs, those are made to do well with desktop workloads but include an iGPU that can handle some amount of workstation workloads as well, Xeon as well as Zen are workstation CPUs, they are made to do well with workstation workloads so they don't have any iGPU to help them with workstation workloads.

What do IGPUs have to do with anything? These are useful for OEM office PCs or notebooks. This is completelly out of topic of this thread or the question you reacted to.

I must apologise but I may stop to react to your posts.
 
Last edited:
  • Like
Reactions: Tlh97 and inf64

TheELF

Diamond Member
Dec 22, 2012
3,973
730
126
There is a lot of evidence, that sales of Intel CPUs for DYI market are down significantly. Have you noticed?
Yes, great for AMD but really so what?! Does it have any influence on the sale numbers of intel? Because all that sales do for intel is go up.
What do IGPUs have to do with anything? These are useful for OEM office PCs or notebooks. This is completelly out of topic of this thread or the question you reacted to.

I must apologise but I may stop to react to your posts.
Spoken like somebody that last used a PC in 2005, GPUs have taken over all the computation heavy workloads decreasing the need for cores that are good at doing the same work since GPUs do it much faster and at much lower cost and lower power.
 
  • Like
Reactions: Dave2150

Jimzz

Diamond Member
Oct 23, 2012
4,399
190
106
I thought that it was reported that Intel actually INCREASED their market-share in CPUs (not sure if DIY was included) recently.


Desktop AMD seems to be pulling ahead of Intel now in some measurements.


Mind you desktop is one of the lowest profit margins for CPUs now. server and laptop are where the money is at. But still to go from 20/80 lose to around 50/50 it a large jump for AMD.


Biggest thing hurting AMD right now is not intel but AMD. They need to get more chips out and recent numbers and ability to buy seem to show they are getting more product out.
 
  • Like
Reactions: Leeea

coercitiv

Diamond Member
Jan 24, 2014
6,196
11,883
136
I thought that it was reported that Intel actually INCREASED their market-share in CPUs (not sure if DIY was included) recently.
Q4 2020 data showed a slight decrease on AMD x86 market share.

1615471795433.png

Relevant tidbits from the article:
"New record highs for total processor revenues were set for both Q4 and 2020, and Q4 set a new record high for quarterly unit shipments, which were more than 125 million units in the quarter," said Dean McCarron of Mercury Research.

Intel obviously captured more of that growth in the quarter than AMD, but it's important to remember that a slight loss of share in the midst of an explosive growth environment doesn't equate to declining sales - AMD grew its processor revenue by 50% last year and posted record financial results for the year. It also shipped more than a million Ryzen 5000 processors in the quarter, according to McCarron research.


PS: I would stay away from market share info built on benchmark usage data, for obvious reasons.
 
  • Like
Reactions: Thibsie

Jimzz

Diamond Member
Oct 23, 2012
4,399
190
106
Q4 2020 data showed a slight decrease on AMD x86 market share.

View attachment 40924

Relevant tidbits from the article:



PS: I would stay away from market share info built on benchmark usage data, for obvious reasons.



Looking at that chart it looks like AMD used its limited wafers mostly for server which is smart. That would explain the scarcity for desktop and mobile until recently.
More money in server than any other.
 

andermans

Member
Sep 11, 2020
151
153
76
Why does the overall market share peak above all the shares in the segments? Wouldn't you expect at least one segment to be above the overall share?
 

Atari2600

Golden Member
Nov 22, 2016
1,409
1,655
136
Not bothered to look it up but I seem to remember JIm Keller stated they succesfully decoupled process from design about half a year before the introduction of sunny/willow/golden cove in december 2018. The need for a backport was already very clear and broadly discussed here (mostly me granted but also others).

I know Intel made much PR bull about this, but that is simply not possible.

You cannot begin to design an efficient microarchitecture without awareness of transistor budgets, in terms of space for routings, size for the reticle (and cost), in terms of switching speeds and power.

Its so intrinsically fundamental to where you make every trade off that the whole idea was awful PR fluff. I'm surprised Keller let his name be attached to such rubbish.
No doubt the actual truth is much more fine-grained; they are likely developing conceptual sub-units independent of process, and taking the knowledge of the emulated performance into conceptual design of a full architecture.



Of course, you could throw together some generic thing using generic rules, but then the amount of performance you are leaving on the table leaves you incredibly vulnerable to a competitor.
 
  • Like
Reactions: Tlh97 and scineram