The AMD Execution Thread [2007 - 2017]

Status
Not open for further replies.
AMD quarterly numbers from WSJ:
Meanwhile, the company saw strong growth in its graphics business. Sales of graphics chips, which are used to process movies, games and other video-intensive applications, jumped 58%, to $427 million, from the same period last year.
The same period last year was quite abnormal due to GFC, will be interesting how well this sustains through the rest of the year.

Looking at the SEC filing here appears graphics is up 40% over last quarter and operating income has become solidly positive at $58m on the $427m compared to $8m on $306m last quarter.
 
Aye, ATI continues to be the one bright spot in AMD's lineup. While both divisions are now showing YoY quarterly gains, the CPU side still saw a YoY yearly loss while the graphics division saw a YoY yearly gain.

AMD seems to be fairly solid with regards to graphics. But on the CPU side they are still at the mercy of Intel. Luckily Intel appears to want to keep AMD's CPU side around as the recent pricing of Core i3/i5 was made to make sure AMD remained the most attractive choice in lower market segments. They could have easily made AMD's CPUs a virtual nonfactor with aggressive pricing.

Regards,
SB
 
AMD quarterly numbers from WSJ:

The same period last year was quite abnormal due to GFC, will be interesting how well this sustains through the rest of the year.

Looking at the SEC filing here appears graphics is up 40% over last quarter and operating income has become solidly positive at $58m on the $427m compared to $8m on $306m last quarter.

Thats excellent, they should at least post as good if not better result this quarter from the increased ramping of higher margin, higher priced DX11 hardware. After Q4 of the financial year its hard to say how things will go relative to Nvidias lineup but I presume their volumes will remain high even if their margins fall at least until Nvidia gets out competing mid/low end graphics solutions as well.
 
So practicly AMD is back to profitable
If you take out the Intel paid money, they're making loss a bit, but this was AFAIK the last quarter GF is reported as part of the company, and taking GF out from the numbers leaves their results positive
 
So practicly AMD is back to profitable
If you take out the Intel paid money, they're making loss a bit, but this was AFAIK the last quarter GF is reported as part of the company, and taking GF out from the numbers leaves their results positive

Wasn't the profit reported > than the money Intel paid in settlement?
 
Aye, ATI continues to be the one bright spot in AMD's lineup. While both divisions are now showing YoY quarterly gains, the CPU side still saw a YoY yearly loss while the graphics division saw a YoY yearly gain.

AMD seems to be fairly solid with regards to graphics. But on the CPU side they are still at the mercy of Intel. Luckily Intel appears to want to keep AMD's CPU side around as the recent pricing of Core i3/i5 was made to make sure AMD remained the most attractive choice in lower market segments. They could have easily made AMD's CPUs a virtual nonfactor with aggressive pricing.

Regards,
SB

It's still a possibility. If Intel feel that AMD are gaining too much market share and their margins become too high, Intel can chop a lot off the low end to squeeze AMD's CPU line.

I hope for AMD's sake that GF100 is kinda crappy, if it comes out and is competitive with 5970 (within 10-20% of performance) it will be very bad for AMD's bottom line and therefore bad for us. I really hope dumping GF works in the long-term, especially once the favourable terms towards AMD end and the board get realistic about who they sell to for how much money.

Either way, I think AMD needs to have at least one unit firing, and if Intel are leaving an open goal on the CPU end I hope AMD make a bit of hay and prepare for difficult times ahead as Intel move down to 32nm in a big way.
 
Nope it was ~100m less than intel apparently paid, which basically means AMD is still in the red.

But Foundry numbers won't be part of the reports anymore this Q, and in last Q they would have been on positive side if both the intel money and foundry numbers would be taken out.
 
It's still a possibility. If Intel feel that AMD are gaining too much market share and their margins become too high, Intel can chop a lot off the low end to squeeze AMD's CPU line.

I hope for AMD's sake that GF100 is kinda crappy, if it comes out and is competitive with 5970 (within 10-20% of performance) it will be very bad for AMD's bottom line and therefore bad for us. I really hope dumping GF works in the long-term, especially once the favourable terms towards AMD end and the board get realistic about who they sell to for how much money.

Either way, I think AMD needs to have at least one unit firing, and if Intel are leaving an open goal on the CPU end I hope AMD make a bit of hay and prepare for difficult times ahead as Intel move down to 32nm in a big way.

For AMDs sake they performance of the GF100 won't be a problem for at least a quarter as getting any significant quantities to market will be troublesome. At this point the question is how long until the GF100 derivatives get to market in quantity, which is where the majority of their volume and margins are made.

On the laptop side, they probably have Q1, Q2 and likely Q3 locked up because they got their products to market in time for the laptop manufacturers and because of the long lead time for laptop makers for their designs. On the desktop side they probably have Q1 and Q2 locked up in their favour because even if they cede the high end, being first to market with volume product DX11 won't be countered for a while by any competing desktop products. That leaves Q3, but if they can get their new architecture out this year it won't be too much of a problem.
 
For AMDs sake they performance of the GF100 won't be a problem for at least a quarter as getting any significant quantities to market will be troublesome. At this point the question is how long until the GF100 derivatives get to market in quantity, which is where the majority of their volume and margins are made.

On the laptop side, they probably have Q1, Q2 and likely Q3 locked up because they got their products to market in time for the laptop manufacturers and because of the long lead time for laptop makers for their designs. On the desktop side they probably have Q1 and Q2 locked up in their favour because even if they cede the high end, being first to market with volume product DX11 won't be countered for a while by any competing desktop products. That leaves Q3, but if they can get their new architecture out this year it won't be too much of a problem.

Aye, how fast Nvidia can get out lower tier cards based on GF100 will be key to how competition is between AMD and Nvidia.

However, things are a bit complicated as Nvidia will use the halo effect of GF100 to market G9x and possible G2xx as the same generation products as GF100 based cards, IE - GT(X) 3xx.

Regards,
SB
 
Nope it was ~100m less than intel apparently paid, which basically means AMD is still in the red.

As was said above, with the GloFo numbers taken out, AMD was slightly positive. Also, if you add in that they likely have a lot of paper losses to trickle out, it may very well be higher. I am going to ask a bunch of analysts about the breakdown in the coming days, they tend to be pretty busy with paying clients in the days after announcements, so I give it a bit.

The reports from all the major players are out on the announcements, if you happen to be in Portland next week, you are welcome to read them.

-Charlie
 
Aye, how fast Nvidia can get out lower tier cards based on GF100 will be key to how competition is between AMD and Nvidia.

However, things are a bit complicated as Nvidia will use the halo effect of GF100 to market G9x and possible G2xx as the same generation products as GF100 based cards, IE - GT(X) 3xx.

Regards,
SB

Honestly, if I see G92 renamed one more time, ONE MORE GODDAMNED TIME! Im going to have a fit of anger so great it'll steam up your glasses from where you're reading it. Luckily I believe they won't be doing it this time so long as their promises about the modularity of Fermi comes true.

I don't believe they will use any GTX 2xx derivatives because the G100 is more efficient per area and hopefully they will finally slay that beast which was the G92.

I will say this though. I believe that ATI are picking up a few extra sales that they otherwise wouldn't because the average enthusiast can actually decipher the characteristics of their lineup from the model names whereas on the Nvidia side of things unless you keep up with things you can't exactly be sure you're not buying the same chip again. Actually replace believe with hope! :(

I wonder how the reviews are going to unfold? Are we going to see Nvidia requiring all review sites to compare against Cypress rather than Hemlock? Are we going to see all review sites required to test a 512SP variant when the initial to market card is say 448SP with 512SP variants being initially as scarce as hens teeth? They hold a lot of sway over 'how' the sites test their products with certain games, and the launch of a new and exciting architecture gives them more room to demand from reviewers certain stipulations. Halo effects are one thing, but first impressions go a long way to developing the aura surrounding the cards and many people may only read one or two reviews and only at the launch of a new architecture, so this is critically important for both Nvidia and AMD to put on a good face to front up with.
 
I wonder how the reviews are going to unfold? Are we going to see Nvidia requiring all review sites to compare against Cypress rather than Hemlock? Are we going to see all review sites required to test a 512SP variant when the initial to market card is say 448SP with 512SP variants being initially as scarce as hens teeth? They hold a lot of sway over 'how' the sites test their products with certain games, and the launch of a new and exciting architecture gives them more room to demand from reviewers certain stipulations. Halo effects are one thing, but first impressions go a long way to developing the aura surrounding the cards and many people may only read one or two reviews and only at the launch of a new architecture, so this is critically important for both Nvidia and AMD to put on a good face to front up with.

Well according to Xman26, it would be a disasterous launch for Nvidia if they allowed G100 to be compared against Hemlock in his explanantion of his post about how 5870 was a disasterous launch because ATI allowed 5870 to be compared against GTX 295 in reviews.

So I'm willing to bet that unless G100 is faster than 5970, that Nvidia will use whatever leverage it can to prevent websites from comparing it to 5970.

Regards,
SB
 
Well according to Xman26, it would be a disasterous launch for Nvidia if they allowed G100 to be compared against Hemlock in his explanantion of his post about how 5870 was a disasterous launch because ATI allowed 5870 to be compared against GTX 295 in reviews.

So I'm willing to bet that unless G100 is faster than 5970, that Nvidia will use whatever leverage it can to prevent websites from comparing it to 5970.

Regards,
SB

Do they have that much leverage? On one hand I can see them doing an editors day and dragging them in, but on the other hand they already have a lot of performance data on the HD 5970 and even without being able to explicitly test the card they can add it to the review anyway. They would have to essentially tell the reviewers to not mention the Hemlock card and then base that on some form of threat or sweetener.
 
Do they have that much leverage? On one hand I can see them doing an editors day and dragging them in, but on the other hand they already have a lot of performance data on the HD 5970 and even without being able to explicitly test the card they can add it to the review anyway. They would have to essentially tell the reviewers to not mention the Hemlock card and then base that on some form of threat or sweetener.

That reply of mine was a bit tongue in cheek. But I have no doubt that Nvidia will have review guidelines. And that they may or may not recommend testing against 5970. And they may or may not use coercion as they have been reported to do in the past, such as refusing review hardware if guidelines aren't followed.

Regards,
SB
 
That reply of mine was a bit tongue in cheek. But I have no doubt that Nvidia will have review guidelines. And that they may or may not recommend testing against 5970. And they may or may not use coercion as they have been reported to do in the past, such as refusing review hardware if guidelines aren't followed.

Regards,
SB


exactly, when NV (or ATI) sends hardware they are often accompanied with guidelines, if you chose not to adhere to them then you most likely can kiss receiving any support (future hardware included) goodbye. Sometimes they'll be brazen enough to pretty much say "You can only test this <X> Software with that <Y> Hardware". People act as though this is some kind of unspoken, underhanded, in the shadows back door dealing but it happens practically every day.
 
For AMDs sake they performance of the GF100 won't be a problem for at least a quarter as getting any significant quantities to market will be troublesome. At this point the question is how long until the GF100 derivatives get to market in quantity, which is where the majority of their volume and margins are made.

On the laptop side, they probably have Q1, Q2 and likely Q3 locked up because they got their products to market in time for the laptop manufacturers and because of the long lead time for laptop makers for their designs. On the desktop side they probably have Q1 and Q2 locked up in their favour because even if they cede the high end, being first to market with volume product DX11 won't be countered for a while by any competing desktop products. That leaves Q3, but if they can get their new architecture out this year it won't be too much of a problem.

Oh indeed, I think even if Fermi blazes the trail and outperforms 5970 it won't have much of an impact as the chip will probably be in short supply for quite a while, probably for the whole if its initial existence as a 40nm part.

The point you make about mainstream parts is important on two fronts, one as you point out they make up most of the dollars spent on discrete graphics and therefore the profit to be earned. Second is the timing, you say Q3 new architecture (as opposed to refresh), it would be very difficult for ATi to introduce a volume product on 28nm on a completely new architecture in Q3 (when TSMC say they go into phase 5 production) even on a refresh it would be quite hard and maintain reasonable yields/costs. If Q3 is their stated goal [for a refresh] it would make sense to do 40nm at massive volumes/high yields (which are apparently available now) and wait until Q1/2 2011 for a new architecture on volume 28nm. Add in that Nvidia are probably hammering TSMC for all of their 28nm availability right now (maybe even willing to pay a lot per wafer, more than ATi at least) because the attraction of moving down to 28nm must be huge for Nv given the heat/size/cost concerns Fermi has.

It wouldn't surprise me if Nvidia's mainstream part never makes an appearance at 40nm being 28/32nm only, with reportedly poor yields at 40nm for GF100, the mainstream won't be that much better, not double anyway. So moving to 28nm and riding out the storm of little availability in Q4 whilst denying ATi 28nm for all 2010 would be a something I would pursue as Nv.

Basically, it looks like GF100 is so uneconomical right now even paying over the odds for 28nm per wafer might be cheaper for Nv per usable die and give better heat/size/cost than 40nm. While Cypress is anything but uneconomical and competing with Nv for 28nm on Nv's terms probably won't suit them, not until volume outstrips supply early in 2011 and TSMC can give lower prices.

Though I wouldn't advocate that ATi sit back on the success of Cypress and the apparent failure of Fermi, it looks like they can rake in a lot of dough between now and Jan 2011 and move to 28nm with a brand new architecture fully tested and ready for mass production while Nv are struggling to make 28nm work on the current architecture.

I don't know, just thinking out aloud...
 
Status
Not open for further replies.
Back
Top