[Computex 2010]AMD to show Fusion tomorrow

Quote:

---

Originally Posted by Clairvoyant129

Yea but anything Intel puts out (no matter what it is) would have been lame for you. :rolleyes:

---

really?
since Intel demo the capability of its SB's GPU I'm expecting it is comparable rather than not comparable.

Quote:

---

Originally Posted by Florinmocanu

I think that is true. The need for really powerful GPUs will remain in the future, the 100$ market and up is here to stay, i cannot imagine how someone could implement for example a GPU the size of 5870 with a CPU onto the same die and keep it under 100W, it's just not possible.

---

Not possible as of today, but who knows what node we will be on (sub 10nm?) in ten years? I'm pretty sure you'd get quadcores and a full 5870 on that node under 100 watts. :up:

Quote:

---

Originally Posted by freeloader

Not possible as of today, but who knows what node we will be on (sub 10nm?) in ten years? I'm pretty sure you'd get quadcores and a full 5870 on that node under 100 watts. :up:

---

hes talking about a mainstream or semi enthusisit card, not just that we will never have something more powerful than a 5870.

the fusion market may be able to offer the same performance as a 200$ cpu + 200$ gpu, but for people who want a 500$ gpu, those will still have to exist, and its very doubtful you will be able to buy a 300mm2 cpu combined with a 400mm2 gpu in a fusion like package.

Quote:

---

Originally Posted by freeloader

Not possible as of today, but who knows what node we will be on (sub 10nm?) in ten years? I'm pretty sure you'd get quadcores and a full 5870 on that node under 100 watts. :up:

---

And by that time, a GPU of 200W on a 10nm process will probably be way more powerfull than the fusion GPU on 10nm.

Fusion is low power, i doubt we will see a fusion product over 100W, that's why we have normal CPUs with lots of cores for. Fusion i think will stay around 60-70W for top-end part.

A GPU of 200-250W is immensely more powerful in games and will always be.

PS: Manicdan explained it pretty decent, 130W CPUs and 100W+ GPUs will always have a market, APUs are designed for low power.

Quote:

---

Originally Posted by Florinmocanu

PS: Manicdan explained it pretty decent, 130W CPUs and 100W+ GPUs will always have a market, APUs are designed for low power.

---

just like 6 years ago you couldnt get a card that was much more than 100W, now you can have over a killowatt of gpu power. looking at the consoles ps1 was 150$, ps2 was 300, and ps3 was 600 new. as time changes people seem to want just more and more and more.

id be scared that 2P will become consumer level in the next decade, lol

Those who are concerned about how a Fusion APU will run games aren't seeing the big picture. APUs are for OpenCL acceleration and other GPGPU applications. In fact, AMD should have done a much better job of emphasizing this fact than showing AVP at low frame rates. 480 shaders on an APU is decent for games, but baller for a GPGPU app. Hell, when the OpenCL GPU3 Folding@Home client shows up, a measly Llano will crank out decent PPD on a laptop.

Quote:

---

Originally Posted by Mechromancer

Those who are concerned about how a Fusion APU will run games aren't seeing the big picture. APUs are for OpenCL acceleration and other GPGPU applications. In fact, AMD should have done a much better job of emphasizing this fact than showing AVP at low frame rates. 480 shaders on an APU is decent for games, but baller for a GPGPU app. Hell, when the OpenCL GPU3 Folding@Home client shows up, a measly Llano will crank out decent PPD on a laptop.

---

I agree there is a nice possibility for GPGPU,but AMD had to show they had a working DX11 Fusion product.And btw,that was Ontario/Bobcat that run the AVP ,with its low power/low clock x86 cores and lowly 80 Stream Processors :)

^^Only 80 stream processors?!?!? Well that isn't as bad as I'd expect performance to be in that game.

Quote:

---

Originally Posted by Manicdan

hes talking about a mainstream or semi enthusisit card, not just that we will never have something more powerful than a 5870.

the fusion market may be able to offer the same performance as a 200$ cpu + 200$ gpu, but for people who want a 500$ gpu, those will still have to exist, and its very doubtful you will be able to buy a 300mm2 cpu combined with a 400mm2 gpu in a fusion like package.

---

Llano wasnt rumored to have a 400 shader gpu ??? so that puts it at 5670 performance depending on clock ...... shrink to the next lowest full node ... and bam 5870 performance is reacheable ... either by increasing shader count or raising clock speed while staying at the same thermal of the 32nm Llano

Quote:

---

Originally Posted by Sn0wm@n

Llano wasnt rumored to have a 400 shader gpu ??? so that puts it at 5670 performance depending on clock ...... shrink to the next lowest full node ... and bam 5870 performance is reacheable ... either by increasing shader count or raising clock speed while staying at the same thermal of the 32nm Llano

---

I wonder what performance will have 5870 with 128 bit bus and DDR3 mem.

Quote:

---

Originally Posted by Manicdan

id be scared that 2P will become consumer level in the next decade, lol

---

1P is the new 2P,
2P is the new 4P.

G34 is turning the market upside-down.
8 real cores for $257.00, 16 for $514.

if you can find it.

:)

Quote:

---

Originally Posted by Sn0wm@n

Llano wasnt rumored to have a 400 shader gpu ??? so that puts it at 5670 performance depending on clock ...... shrink to the next lowest full node ... and bam 5870 performance is reacheable ... either by increasing shader count or raising clock speed while staying at the same thermal of the 32nm Llano

---

you might be a little off on the expectations. if we can get 480SPs on Llano at 32nm, then at the next step, 22nm, (or is it 24? idk...) we can expect 50% more, with maybe 10% more clock rate at the same power. so its 800SPs if were lucky, and clocking probably 1ghz, sounds like an OCed 5770, or maybe up to a 5830. the next one will probably be well within reach of a 5870 though. so that means <1 year till this, and then 2 years for each node after, so just under 5 years away and u can have a 5870 with probably an 8 core BD chip under 100W.

kinda scary what laptops and nettops will be able to do in 5 years.

Quote:

---

Originally Posted by wuttz

1P is the new 2P,
2P is the new 4P.

G34 is turning the market upside-down.
8 real cores for $257.00, 16 for $514.

if you can find it.

:)

---

lol, its limit 1 per customer for a 2P chip

Quote:

---

Originally Posted by Manicdan

you might be a little off on the expectations. if we can get 480SPs on Llano at 32nm, then at the next step, 22nm, (or is it 24? idk...) we can expect 50% more, with maybe 10% more clock rate at the same power. so its 800SPs if were lucky, and clocking probably 1ghz, sounds like an OCed 5770, or maybe up to a 5830. the next one will probably be well within reach of a 5870 though. so that means <1 year till this, and then 2 years for each node after, so just under 5 years away and u can have a 5870 with probably an 8 core BD chip under 100W.

kinda scary what laptops and nettops will be able to do in 5 years.

---

the 22nm will reduce the die size by close to 40% - 50% cut power down to 30% .... so using that die space to add more shader and use the norther island shader design wouldnt be a bad idea on the 22nm node gaining another 20% more performance ... and bam 5870 performance right there .... on a decent tdp

50% more shaders, using 30% less power, with 20% ipc gains over Llano will = 5870? very doubtful.

if Llano is 1x, then 1 * 1.5 for space, and * 1.2 for IPC = 1.8x. so a 22nm version of Llano is 1.8x better than 32nm version. so by that standard, the current Llano has to be better than a 5770. and even if it is, with only a 30% power drop and being 50% bigger, your going to have a much higher TDP at the same clocks, effectively reducing how high you can clock it, say 800mhz vs 700mhz with a .1v drop.

Quote:

---

Originally Posted by Manicdan

lol, its limit 1 per customer for a 2P chip

---

Now there's the laugh of the day :rofl:

Quote:

---

Originally Posted by Manicdan

lol, its limit 1 per customer for a 2P chip

---

news flash! it isn't 2P only, that chip goes across the board from 1P to 4P.
one reason it's 1/customer only is probably the tight demand.

the other reason is this. did i mention G34 is blurring server market segements?

Quote:

---

Originally Posted by Sn0wm@n

the 22nm will reduce the die size by close to 40% - 50% cut power down to 30% .... so using that die space to add more shader and use the norther island shader design wouldnt be a bad idea on the 22nm node gaining another 20% more performance ... and bam 5870 performance right there .... on a decent tdp

---

different types of logic scale differently and there is a huge difference in density. most of the gpu is SRAM and datapaths which will scale very well. think 2x more logic from 32nm to 22nm for the gpu. hopefully Glofo will quit using SOI past 32nm and density will increase even more.

Quote:

---

Originally Posted by wuttz

news flash! it isn't 2P only, that chip goes across the board from 1P to 4P.
one reason it's 1/customer only is probably the tight demand.

the other reason is this. did i mention G34 is blurring server market segements?

---

wrong, it can go all the way up to 8p in one rig with cpu extension board.

But the cost for only a 4 socket motherboard and 2x 2socket cpu extension board would cost more than 8 opteron 6128 ( the 300$ 8 core 2.0ghz cpu). But hey, you would have a 64 core rig !!!!
And the n you can upgrade to lowest bulldozer server cpu which will be 12 core and have 96 core or go batcrazy and get the 16 core one and have 128 core.

Quote:

---

Originally Posted by Chumbucket843

hopefully Glofo will quit using SOI past 32nm and density will increase even more.

---

So SOI has worse transistor density?

Quote:

---

Originally Posted by snoro

wrong, it can go all the way up to 8p in one rig with cpu extension board.

But the cost for only a 4 socket motherboard and 2x 2socket cpu extension board would cost more than 8 opteron 6128 ( the 300$ 8 core 2.0ghz cpu). But hey, you would have a 64 core rig !!!!
And the n you can upgrade to lowest bulldozer server cpu which will be 12 core and have 96 core or go batcrazy and get the 16 core one and have 128 core.

---

What is this "CPU extension board" that you speak of?

my bad, i thought one of the 2 cpu supermicro motherboard pcie express 16x slot was a proprietary link to act as a extension bridge.

SM 2 socket g34 motherboard with one very strange pcie 16x slot placement

Quote:

---

Originally Posted by FischOderAal

So SOI has worse transistor density?

---

yup. their are tradeoffs for SOI. you'll get higher clocks but there are numerous disadvantages. for instance the insulator is also an insulator of thermal energy which means it will get hot unlike a normal bulk process where the heat can diffuse into the substrate. it's also more complex and costs more. i think amd used it b/c they collaborated with ibm on their 45nm node and probably some other tech.

L2-cache is shared ?

http://pc.watch.impress.co.jp/docs/c...04_371991.html

http://pc.watch.impress.co.jp/img/pc...371/991/10.jpg

Quote:

---

Originally Posted by superrugal

L2-cache is shared ?
http://pc.watch.impress.co.jp/docs/c...04_371991.html

---

Rather unlikely in my opinion....

From the same page, AMD's Ontario wafer:
http://pc.watch.impress.co.jp/img/pc...371/991/33.jpg

AMD's dual Bobcat core + DX11 GPU Ontario is remarkably small: ~75mm^2
(>900 dies per wafer) which is smaller as the dual core atom based D510
Pineview which has 87mm^2

Intel dual core Atom D510: http://ark.intel.com/Product.aspx?id=43098
http://media.bestofmicro.com/dual-co...3-233967-3.jpg


Regards, Hans

Hi guys, are we sure that ontario is 100% cpu+gpu in the same die or maybe is a clarkdale kind of accommodation? because those dies are awfully tiny, maybe they are only the cpus? and the gpus are made at tsmc 40nm?

I mean if amd can pull off that kind of performance in 75mm? that would be extremely impressive

They're 32nm yeah? so I doubt at ~75mm it'd be just the CPU.

After all, 512KB L2 Dual core K10 @ 32nm would be around that small /smaller.

-edit, and another thing to ponder. Has AMD been able to shrink the memory controller more successfully on 32nm? and what about Cache Density?

Both taking up a huge amount of space on Regor and propus reletive to the cores themselves. There were some figures around last yr that showed promise for Cache Density at least

Quote:

---

Originally Posted by mAJORD

They're 32nm yeah? so I doubt at ~75mm it'd be just the CPU.

After all, 512KB L2 Dual core K10 @ 32nm would be around that small /smaller.

---

ontario is 40nm. not much is known but bobcat is a fully synthesized design so dont expect amazing efficiency. they'll certainly be able to add more sram with tsmc though. im pretty sure tsmc also has better support for SoC's too.

Quote:

---

Originally Posted by Hans de Vries

From the same page, AMD's Ontario wafer:
http://pc.watch.impress.co.jp/img/pc...371/991/33.jpg

---

BTW, the case looks like those being used by GF (in some cases):
http://www.fudzilla.com/images/stori...s/gfwafer2.jpg
(http://www.fudzilla.com/content/view/16343/1/)

Quote:

---

Originally Posted by Dresdenboy

BTW, the case looks like those being used by GF (in some cases):
http://www.fudzilla.com/images/stori...s/gfwafer2.jpg
(http://www.fudzilla.com/content/view/16343/1/)

---

The wafer showed by FUD was 32nm SOI.

Enclosure can be standard across many fabs. But I don't know so the possibility is there!
I wonder how late GloFo 40nm bulk will compare to TSMC 40nm HP bulk.
GloFo is late to the party and first bulk development since long time but it might just be better than competition.

there is no high power 40nm offering from global foundries and from their roadmaps it doesnt look like they will have one. all of their highest performance processes will be using SOI so that makes it hard to compare them to tsmc's nodes. although their other nodes are still bulk.

Quote:

---

Originally Posted by superrugal

The wafer showed by FUD was 32nm SOI.

---

That's correct, but this was not my point. I mentioned the case/enclosure.

Quote:

---

Originally Posted by Lightman

Enclosure can be standard across many fabs. But I don't know so the possibility is there!
I wonder how late GloFo 40nm bulk will compare to TSMC 40nm HP bulk.
GloFo is late to the party and first bulk development since long time but it might just be better than competition.

---

Only TSMC is interesting here. I was looking for photos of TSMC wafer presentations. If there was an enclosure, it had a rather simplistic design.

BTW, did Rick say, that this is a 40nm wafer? To my knowledge, this is just speculation, because he mentioned the low power Fusion processor and netbooks. Also the performance looks like what we might expect from Ontario. OTOH 40nm LP should be ready at GF:
http://pc.watch.impress.co.jp/img/pc...l/017.jpg.html

Quote:

---

Originally Posted by Chumbucket843

there is no high power 40nm offering from global foundries and from their roadmaps it doesnt look like they will have one. all of their highest performance processes will be using SOI so that makes it hard to compare them to tsmc's nodes. although their other nodes are still bulk.

---

40 nm HP is not needed. 40 nm LP is the appropriate process as linked above and here:
http://pc.watch.impress.co.jp/img/pc...ml/10.jpg.html

Quote:

---

Originally Posted by railer

terrace215 you are not seriously implying that AMD has some hidden wood screws here? While there is some validity to your questions and we would like to see the system board Fusion is running on and its TDP, its hard to imagine that company like AMD would only have bluff on their hands.

---

I submit the AMD "Barcelona QC running at 3 GHz" demo as counterpoint. AMD has, in the past, been willing to purposely create... let's say, "overly optimistic"... impressions.

But anyhow, more interesting may be "what's up with the Llano no-show in terms of a public demo?"

And the ongoing complete public silence on the status of Bulldozer, supposedly "for competitive reasons" (Yeah, right!)

Well, maybe IDF will shake some info out of AMD in September, if the analysts don't get it in July.

Quote:

---

Originally Posted by terrace215

I submit the AMD "Barcelona QC running at 3 GHz" demo as counterpoint. AMD has, in the past, been willing to purposely create... let's say, "overly optimistic"... impressions.

But anyhow, more interesting may be "what's up with the Llano no-show in terms of a public demo?"

And the ongoing complete public silence on the status of Bulldozer, supposedly "for competitive reasons" (Yeah, right!)

Well, maybe IDF will shake some info out of AMD in September, if the analysts don't get it in July.

---

Haven't you said all of this before?

Quote:

---

Originally Posted by terrace215

I submit the AMD "Barcelona QC running at 3 GHz" demo as counterpoint. AMD has, in the past, been willing to purposely create... let's say, "overly optimistic"... impressions.

But anyhow, more interesting may be "what's up with the Llano no-show in terms of a public demo?"

And the ongoing complete public silence on the status of Bulldozer, supposedly "for competitive reasons" (Yeah, right!)

Well, maybe IDF will shake some info out of AMD in September, if the analysts don't get it in July.

---

haven't seen a sandybridge board either. :p

Given amds recent track record im sure they will do fine. I think they learned their lesson with phenom and have slowly but surely gotten themselves back on track.

How hard is it to understand? From the PR point of view if you have slower part which makes everyone to say wow on your hand why would you want to show faster part. if slower product is already drawing so.much attention and interest, what is the point of showing the faster stuff publicly. We know that they showed Llano behind the closed doors and we know that partners already have samples. If I had a low power product so impressive as Ontario demo showed, I would also keep high power stuff closer to myself and let the competition think: What the f**k are they going to do to us with performance parts if low powered stuff is doing things like that??

Im buying 300 shares of AMD sometime this week:yepp::D

Quote:

---

Originally Posted by Dresdenboy

40 nm HP is not needed. 40 nm LP is the appropriate process as linked above and here:
http://pc.watch.impress.co.jp/img/pc...ml/10.jpg.html

---

why not? im not specifically referring to ontario or any specific processor, just the nodes in general. i still think tsmc is very competitive with glofo in every way. sure people put tsmc in a bad light for their yields but that wouldnt be such a problem if ATi and nvidia were not so "aggressive" for lack of a better word.

Quote:

---

Originally Posted by slaveondope

Im buying 300 shares of AMD sometime this week:yepp::D

---

why? is this fusion so much good?

Quote:

---

Originally Posted by Kal El

why? is this fusion so much good?

---

Not just "Fusion" but the up coming ATI 6 series, Bulldozer, Llano, and Ontario. Im expecting a good year for AMD in 2011 and I figure I might as well ride the wave. *crosses fingers* Then bail out when Intel fights back;)

Quote:

---

Originally Posted by jakefalcons

Given amds recent track record im sure they will do fine. I think they learned their lesson with phenom and have slowly but surely gotten themselves back on track.

---

Every track record is meant to be broken sometime. But I agree, AMD seems to have learned it's lessons and changed for the better.

Quote:

---

Originally Posted by slaveondope

Not just "Fusion" but the up coming ATI 6 series, Bulldozer, Llano, and Ontario. Im expecting a good year for AMD in 2011 and I figure I might as well ride the wave. *crosses fingers* Then bail out when Intel fights back;)

---

Quote:

---

Originally Posted by FischOderAal

Every track record is meant to be broken sometime. But I agree, AMD seems to have learned it's lessons and changed for the better.

---

i do not understand? from what i have learned in my correspondence with OhNoes!, amd has not been more better than Intel, and most likely will not be? does amd have better now with fusion?

Past performance isn't necessarily an indicator of future performance. Any company can mess up - AMD, Intel, Nvidia, Microsoft, IBM, etc.

Quote:

---

Originally Posted by Kal El

i do not understand? from what i have learned in my correspondence with OhNoes!, amd has not been more better than Intel, and most likely will not be? does amd have better now with fusion?

---

Intel has the faster chips right now. It hasn't always been the case in the past and won't necessarily be the case in the future. Anyone who is claiming to know how Bulldozer and Sandy Bridge will compare to each other is probably displaying a case of wishful thinking, extreme speculation, etc.

Quote:

---

Originally Posted by terrace215

I submit the AMD "Barcelona QC running at 3 GHz" demo as counterpoint. AMD has, in the past, been willing to purposely create... let's say, "overly optimistic"... impressions.

But anyhow, more interesting may be "what's up with the Llano no-show in terms of a public demo?"

And the ongoing complete public silence on the status of Bulldozer, supposedly "for competitive reasons" (Yeah, right!)

Well, maybe IDF will shake some info out of AMD in September, if the analysts don't get it in July.

---

recently AMD is showing something realistic rather than over optimistic
Look at the launch of Istanbul, Thuban and Deneb;
your link to the barcelona post is like fanboi act:rofl::rofl:

avp demo is very impressive. i hope AMD has done a monster. We need PC that can run 10h of 3D.

Quote:

---

Originally Posted by Kal El

i do not understand? from what i have learned in my correspondence with OhNoes!, amd has not been more better than Intel, and most likely will not be? does amd have better now with fusion?

---

well the problem is what you believe when you email with some people and certainly when you mention this:


in my correspondence with OhNoes! :rofl::rofl:

So we've learned so far:
1. Intel fanboys hate AMD for no particular reason.
2. Ontario and Llano might be good chips for their intended purpose.

Quote:

---

Originally Posted by MrMojoZ

So we've learned so far:
1. Intel fanboys hate AMD for no particular reason.
2. Ontario and Llano might be good chips for their intended purpose.

---

3. 2010 is feeling so very long of a year

Quote:

---

Originally Posted by duploxxx

well the problem is what you believe when you email with some people and certainly when you mention this:


in my correspondence with OhNoes! :rofl::rofl:

---

Hold your horses. I don't even know this guy, nor have I received any PMs from him. An admin could probably confirm this; heck, I'm sure he can confirm it himself. I've had no "correspondence" with this guy on any matter or topic. Period. Did it even occur to you this guy could simply be using the wrong word since it appears he is not a Native Speaker of English? :stick:

Secondly, I'm glad he seems to have read my posts as they simply state facts, and not some contrived bs. Facts don't lie. :cool:

Quote:

---

Originally Posted by MrMojoZ

So we've learned so far:
1. Intel fanboys hate AMD for no particular reason.
2. Ontario and Llano might be good chips for their intended purpose.

---

If I am a fanboy for stating the obvious (read fact) then I vy for the biggest fanboy on XS. I am not going to be ridiculed into making any false claims about a product. Sorry, gullibility and jumping to conclusions are not particularly two of my fortes.

PS: I await what Kal El has to say about this. Hopefully, he's not one of the AMD bots out to draw blood.:ROTF:

i am sorry, i have read posts...i use translate from Google for some words, it must have been wrong. please i am no hurt, i am sorry. i havened read posts (i was looking for better phrasing before), i thought your posts stood out. please i am no hurt.

:):):):)

i have made error and i punish myself.

Nice pretending going on there

Actually since he is from Krypton he may be using some form of automatic translation device :D .

check my sig and that hardware is 1-2 years same.
I got no need for new

If any amd cpu can beat mine cpu that im using, for reasonable price, then im buying amd in like 4 years from now(q9550 still lives on top)
and I havent start to to overclocking it :)