Text Size

AMD Fusion, Bobcat, Bulldozer

Charlie says that 28nm Bobcate follow up has taped out...

Discussion about AMD's upcoming CPU's and APU's

Re: Charlie says that 28nm Bobcate follow up has taped out..

Postby BaronMatrix » Sun Apr 17, 2011 10:58 pm

Reiuji Utsuho wrote:
Kedas wrote:http://www.xbitlabs.com/articles/cpu/display/core-i3-2100t_9.html#sect0

Why is the Core i3-2100T better in game tests than the E-350.
It uses even less energy in 1080p playback.
E-350 certainly isn't going to be better in CPU performance but also a slower GPU/UVD3 for about the same power.

edit:
i3-2100T TDP is 35W
E-350 TDP is 18W
It looks like the CPU is the bottleneck for the GPU of the E-350.


Ordinary Kill-a-watt measurement makes no sense in low power platforms. Most of the power is lost during conversion. Linus tech tips measured 32W for a C-50 tablet, but that thing runs 6 hours on a 28Whr battery.

Low TDP does not make E-350 a power efficient processor, just because its raw performance is too weak. It's not surprising E-350 consumes more power than i3 2100T in some of the tests: it has too little compute power, so it needs to be highly loaded in order to complete that task; i3 has a lot to spare, so the workload is very low. This is also the problem for Atom: a weak processor need to work continuously to finish the task done very quickly on a faster processor, which can be idle for most of the time.

Sandybridge is faster and more efficient than Zactate, that's a given because it's 2 times larger, built on 32nm and cost you two times the money. There is no free lunch here. But being 4 times faster than E-350 does not make it a good processor. Llano will have more-much, much more compute power and should consume similar power and it will crush SNB with ease. On the other hand, being only one fourth as fast as Sandybridge does not make Zactate a poor processor: Atom is too slow to do anything, but Zactate can do whatever you throw at it. This relative advantage makes Zactate very competitive in low-end desktop and netbooks; Sandybridge does not aim at this market segment at all.

The numbers themselves have no problem. But the article itself is the problem. It's the same thing as comparing a Chevy with Mercedes.




The XBit review admits that there is a better MOBO to use that does use a few watts less than the i3. Plus Zacate was built for low power entry level perf and shouldn't be compared to an i3. The low voltage Llanos will fit that niche. Perhaps when the 28nm happens the clocks will rise enough from 1.6GHz to make a better competitor for those chips but that's still not the market it addresses. Especially in price.
MSI 790GX X4 940 - @ 3GHz - AC64 HSF - HD 4870 512MB - 8GB DDR2 - 1TB HDD - BluRay - TX650 W - Antec 300 - 2X ACER 23"
HP DV7-3065 - M600 - 17" LCD - BD-Rom - HDMI\eSATA - .5TB HDD - HD4200
User avatar
BaronMatrix
K8 Athlon 64 (San Diego) Expert Boarder
K8 Athlon 64 (San Diego) Expert Boarder
 
Posts: 1949
Joined: Mon Jan 29, 2007 3:13 am

Re: Charlie says that 28nm Bobcate follow up has taped out..

Postby abinstein » Sun Apr 17, 2011 11:06 pm

Smartidiot89 wrote:Precisely, people are to quick to call bullsh*t on a review without even thinking about the most important factor = Price. Personally I see nothing wrong in this review which is being discussed but I guess that's because I am using an objective eye and is looking at the cost of what is being reviewed. Brazos is a great platform, and lets just hope Llano and Bulldozer is enough to raise the bar? ;)

I see you're trying to find good excuse for people being evolutionary backward.

If a review comparing products with drastically different cost intended for completely different markets is not a bad review, then we must all be monkeys.

Maybe they should review bananas instead. :roll:

(BTW, without an AMD gpu, sandy bridge will fail on any app with gpgpu acceleration. If that's not what "enthusiasts" should be looking at, then we're really all monkeys.)
abinstein
K8 Opteron (SledgeHammer) Moderator
K8 Opteron (SledgeHammer) Moderator
 
Posts: 7176
Joined: Sat Oct 30, 2004 9:49 pm

Re: Charlie says that 28nm Bobcate follow up has taped out..

Postby Smartidiot89 » Sun Apr 17, 2011 11:20 pm

abinstein wrote:
Smartidiot89 wrote:Precisely, people are to quick to call bullsh*t on a review without even thinking about the most important factor = Price. Personally I see nothing wrong in this review which is being discussed but I guess that's because I am using an objective eye and is looking at the cost of what is being reviewed. Brazos is a great platform, and lets just hope Llano and Bulldozer is enough to raise the bar? ;)

I see you're trying to find good excuse for people being evolutionary backward.

If a review comparing products with drastically different cost intended for completely different markets is not a bad review, then we must all be monkeys.

Maybe they should review bananas instead. :roll:

(BTW, without an AMD gpu, sandy bridge will fail on any app with gpgpu acceleration. If that's not what "enthusiasts" should be looking at, then we're really all monkeys.)

The only monkey is the ones who can't look at a review with an objective eye and in this case take price into account.
Smartidiot89
 

Re: Charlie says that 28nm Bobcate follow up has taped out..

Postby abinstein » Sun Apr 17, 2011 11:22 pm

BaronMatrix wrote:
AussieFX wrote:
abinstein wrote: since netbook is the first PC type that's going to be replaced by tablets.

I'm not so sure about that. I predict that when tablets start breaking (which they will) people will go back to netbooks for the added protection.
And a quadcore netbook would be far better from a productivity point of view than any single/dualcore tablet.

I think tablets will fall into a kids gaming unit and not much more.



Totally agree. I have yet to find a real use for something that big when I have a 4.3" phone that is more than enough. And the phone will run "desktop" apps. (WP7 Dell Venue Pro). But I see Krishna being a 2.5GHz dual - which if history is any indicator, most modern archs don't really "kick in" until 2GHz - and a 2GHz quad which should sit under the lowest Llano or Trinity.


I'll just ask you two to bookmark this page and come back a year later.
abinstein
K8 Opteron (SledgeHammer) Moderator
K8 Opteron (SledgeHammer) Moderator
 
Posts: 7176
Joined: Sat Oct 30, 2004 9:49 pm

Re: Charlie says that 28nm Bobcate follow up has taped out..

Postby abinstein » Sun Apr 17, 2011 11:24 pm

Smartidiot89 wrote:
abinstein wrote:
Smartidiot89 wrote:Precisely, people are to quick to call bullsh*t on a review without even thinking about the most important factor = Price. Personally I see nothing wrong in this review which is being discussed but I guess that's because I am using an objective eye and is looking at the cost of what is being reviewed. Brazos is a great platform, and lets just hope Llano and Bulldozer is enough to raise the bar? ;)

I see you're trying to find good excuse for people being evolutionary backward.

If a review comparing products with drastically different cost intended for completely different markets is not a bad review, then we must all be monkeys.

Maybe they should review bananas instead. :roll:

(BTW, without an AMD gpu, sandy bridge will fail on any app with gpgpu acceleration. If that's not what "enthusiasts" should be looking at, then we're really all monkeys.)

The only monkey is the ones who can't look at a review with an objective eye and in this case take price into account.


Here's your banana: how does the review make sense when it compares products with completely different price ranges?

The cheaper one has gpgpu, the more expensive one doesn't. Did the review look at that?

The cheaper one is power efficient, the more expensive one is not. Did the review compare them fairly?

Did the review use proper workload (who's going to game on such rigs)?

Talking about objectivity. The objective view of the review is that its comparison is senseless. I find it surprising that you like this review, because by your own words whoever can't objectively find the problem in this review are monkeys, and I don't think monkeys speak English like you do.
abinstein
K8 Opteron (SledgeHammer) Moderator
K8 Opteron (SledgeHammer) Moderator
 
Posts: 7176
Joined: Sat Oct 30, 2004 9:49 pm

Re: Charlie says that 28nm Bobcate follow up has taped out..

Postby dm7000s » Mon Apr 18, 2011 12:34 am

BaronMatrix wrote:
The XBit review admits that there is a better MOBO to use that does use a few watts less than the i3.


Ah yes.... there is. But those results couldn't be included in their pre-cooked conclusion in any meaningful way.

XBit Labs: MSI E350
Idle: 7.3W (SB 9.7W) -> SandyBridge: +32%
CPU-Load: 15.8W (SB 23.9W) -> SandyBridge: +51%
GPU-Load: 17.5W (SB 27.7W) -> SandyBridge: +58%
CPU-GPU: 22.1W (SB 31.2W) -> SandyBridge: +41%
1080p: 14.9W (SB 17.7W) -> SandyBridge: +19%

So what do we have here? Now intel needs to position their twice as large (149 mm²), carefully selected -low yield- toned down higher cost top product, against AMD's cheap-ass machine-synthesized (designed to be stamped out as quickly and as cheaply as possible) to become *almost* competitive?

As in: We charge as much for the processor as AMD does for the whole platform? ROFL!

Wow.. what competitiveness does 2100T bring!
-Higher power consumption .... check!
-Non-distinguishable (platform intended) performance .... check!
-Lack of age-old features (DX11 etc.) .... check!
-23.976 fps bug still present .... check!
-Ridiculously higher price .... check!

This xbitlabs "review", intended to bring the 2100T as the holy-grail, does in fact the exact opposite! Showing that the 2100T is something to stay far away from.
Image
User avatar
dm7000s
K6-2 Fresh Boarder
K6-2 Fresh Boarder
 
Posts: 218
Joined: Wed May 07, 2008 11:24 pm

Re: Charlie says that 28nm Bobcate follow up has taped out..

Postby Smartidiot89 » Mon Apr 18, 2011 8:23 am

abinstein wrote:Here's your banana: how does the review make sense when it compares products with completely different price ranges?

The cheaper one has gpgpu, the more expensive one doesn't. Did the review look at that?

The cheaper one is power efficient, the more expensive one is not. Did the review compare them fairly?

Did the review use proper workload (who's going to game on such rigs)?

Talking about objectivity. The objective view of the review is that its comparison is senseless. I find it surprising that you like this review, because by your own words whoever can't objectively find the problem in this review are monkeys, and I don't think monkeys speak English like you do.

It shows what people are looking for, and for the tests they use the results looks accurate. That's good enough for me. I can't say one product is better than the other based on one review I need more sources to do so, in this case I say both products have advantages and disadvantages over each other while one of them is much cheaper.
Smartidiot89
 

Re: Charlie says that 28nm Bobcate follow up has taped out..

Postby Am486DX100 » Mon Apr 18, 2011 2:07 pm

"•Power supply: Tagan TG880-U33II (880 W)."

Looks like overkill for using this power supply for a Mini-ITX board. My current system would make better use of it than the Mini-ITX.
Ah, the old days before there was this fashion for having a solo career.

Squeak?

Yes, once there were five horsemen. But you know;there's always a row. Creative disagreements, rooms being trashed, and things said that perhaps should not have been said.
User avatar
Am486DX100
K8 Athlon 64 (Venice) Expert Boarder
K8 Athlon 64 (Venice) Expert Boarder
 
Posts: 1675
Joined: Fri May 26, 2006 7:17 pm
Location: Ahhhh..... I'm @rting around!!!

Re: Charlie says that 28nm Bobcate follow up has taped out..

Postby mmarq » Mon Apr 18, 2011 2:28 pm

Reiuji Utsuho wrote:Ordinary Kill-a-watt measurement makes no sense in low power platforms. Most of the power is lost during conversion. Linus tech tips measured 32W for a C-50 tablet, but that thing runs 6 hours on a 28Whr battery.


Reiuji Utsuho wrote:Low TDP does not make E-350 a power efficient processor, just because its raw performance is too weak. It's not surprising E-350 consumes more power than i3 2100T in some of the tests:


Where do you guys came from ??.. its obvious that i3 has more performance than a zacate, maybe its performance its "weak", but no where is i3 double the performance of Zacate(even on the biased xbitlab test).. yet it might consume almost double the power, as a "WHOLE" system.. so you talk about performance/watt or performance/$ !?... (Intel loses in either case!)

Your conclusions make me :lol: ... when so obvious contradictions doesn't make you think.. and pause ??

I have the "Washington Monument" to sell if you are interested "Reiuji Utsuho" !?

From your own post isn't obvious that a whole Zacate system consumes less than a single Intel CPU.. which BTW, as seen before, might very well have power management tricks making those measurements dubious and at a low level standard, because of optimized benchmarketoses even the ones for measuring power, and so "avoids" NORMAL SPIKES that might put it way above in average power consumption !??...

Show me "measurements at the wall".. on tests that might consider multitasking... and don't be surprised(at least i wont) to see that a single Intel i3 CPU SYSTEM consumes almost the DOUBLE of a whole Zacate SYSTEM ...

[EDIT: i might even consider low end desktop... if you put a whole lot of "low end" in it.. but Zacate its clearly for ultramobile/embedded systems.. which SB is not(CLEARLY).. so again the problem is ORANGES vs Apples.. and so this thread is derailed about an obvious biased test from a "site" that has made ppl at this site in the past, scream at so many obvious "twists" ]
mmarq
K8 Athlon 64 (Orleans) Expert Boarder
K8 Athlon 64 (Orleans) Expert Boarder
 
Posts: 2337
Joined: Sat Jul 14, 2007 4:31 am

Re: Charlie says that 28nm Bobcate follow up has taped out..

Postby mmarq » Mon Apr 18, 2011 3:06 pm

more!.. i see vindicated my rants about such embedded/ultramobile offerings having chipsets capables of x16 PCI slots.. or even PCI ALTOGHETHER... AMD is going to suffer by Intel trying to squeeze every penny out of morons pockets.. when a proper chipset would avoid this tricks(isn't HT better for even low power=> its power managed?).. the game of OEM PCI demand can have a "carrot" with the wide spread perspective and more profits, but in the end it serves best the old master... in the end this "probable" RUSH to Wichita and Krishna might be to fill that gap!..( NOW its back on topic!)
mmarq
K8 Athlon 64 (Orleans) Expert Boarder
K8 Athlon 64 (Orleans) Expert Boarder
 
Posts: 2337
Joined: Sat Jul 14, 2007 4:31 am

Re: Charlie says that 28nm Bobcate follow up has taped out..

Postby mmarq » Mon Apr 18, 2011 3:23 pm

BaronMatrix wrote: The interesting thing will be Trinity. From the roadmap, it should tape out soon. It is 32nm. The interesting thing about Krishna is that AMD showed an ARM chip on SOI that showed 40% power savings so I would think that Krishna would aim towards SOI at least at GF. It should be just an optical shrink of 32nm. That would let clocks go even higher than BULK.


It might happen that Trinity is even for HPC Workstation/server ?.. well! maybe not Trinity... but it clearly can in a very advantageous way, IF when GPGPU takes IN very strong!... talking about a possible MCM of APUs, with 4 to 8 CPU cores + GPGPU per die!... and so making a kind of G34 type of socket with 8 to 16CPU cores and 2 GPGPUs...

Where you guys think trinity would land on!

Image

If not the whole thing of that Phase 3, at least having a more evolved "hardware scheduler" would put it clearly and ostensibly in GPGPU territory... (by byby blackbird!.. i mean Intel! :lol: ) :mrgreen:

[ EDIT: at least GPU with high level language support seems obvious OpenCL 2.0.. and AMD seems to be leading the thing http://sa09.idav.ucdavis.edu/docs/SA09- ... erview.pdf ]
mmarq
K8 Athlon 64 (Orleans) Expert Boarder
K8 Athlon 64 (Orleans) Expert Boarder
 
Posts: 2337
Joined: Sat Jul 14, 2007 4:31 am

Re: Charlie says that 28nm Bobcate follow up has taped out..

Postby Smartidiot89 » Mon Apr 18, 2011 7:30 pm

mmarq wrote:
BaronMatrix wrote: The interesting thing will be Trinity. From the roadmap, it should tape out soon. It is 32nm. The interesting thing about Krishna is that AMD showed an ARM chip on SOI that showed 40% power savings so I would think that Krishna would aim towards SOI at least at GF. It should be just an optical shrink of 32nm. That would let clocks go even higher than BULK.


It might happen that Trinity is even for HPC Workstation/server ?.. well! maybe not Trinity... but it clearly can in a very advantageous way, IF when GPGPU takes IN very strong!... talking about a possible MCM of APUs, with 4 to 8 CPU cores + GPGPU per die!... and so making a kind of G34 type of socket with 8 to 16CPU cores and 2 GPGPUs...

Where you guys think trinity would land on!

Image

If not the whole thing of that Phase 3, at least having a more evolved "hardware scheduler" would put it clearly and ostensibly in GPGPU territory... (by byby blackbird!.. i mean Intel! :lol: ) :mrgreen:

[ EDIT: at least GPU with high level language support seems obvious OpenCL 2.0.. and AMD seems to be leading the thing http://sa09.idav.ucdavis.edu/docs/SA09- ... erview.pdf ]

Nice image there! Microsoft said in a presentation that Windows 8 will utilize GPGPU "wherever possible" in their operating system so I am sure AMD will gain a huge advantage there once Windows 8 is out. Also makes perfect sense since Windows 8 supposedly will be able to get "cut-down" for use in Tablets :mrgreen:
Smartidiot89
 

Re: Charlie says that 28nm Bobcate follow up has taped out..

Postby AussieFX » Tue Apr 19, 2011 4:51 am

mmarq wrote:
BaronMatrix wrote: The interesting thing will be Trinity. From the roadmap, it should tape out soon. It is 32nm. The interesting thing about Krishna is that AMD showed an ARM chip on SOI that showed 40% power savings so I would think that Krishna would aim towards SOI at least at GF. It should be just an optical shrink of 32nm. That would let clocks go even higher than BULK.


It might happen that Trinity is even for HPC Workstation/server ?.. well! maybe not Trinity... but it clearly can in a very advantageous way, IF when GPGPU takes IN very strong!... talking about a possible MCM of APUs, with 4 to 8 CPU cores + GPGPU per die!... and so making a kind of G34 type of socket with 8 to 16CPU cores and 2 GPGPUs...

I think Llano had all it's outside ht links removed to save power/space so it's out but BDv2 could certainly fulfil that role.
Sent from my flippy phone thingy using TAPATALK HD_2016.1


Image
Nikon D7000 / Nikon D5000
User avatar
AussieFX
K8 Opteron (SledgeHammer) Moderator
K8 Opteron (SledgeHammer) Moderator
 
Posts: 7660
Joined: Fri May 11, 2007 1:50 pm
Location: I wish I knew...

Re: Charlie says that 28nm Bobcate follow up has taped out..

Postby mmarq » Tue Apr 19, 2011 7:01 pm

Well! we don't know exactly if Llano has ALL HT links removed.. that is a pretty good guess... but certainly it wasn't to save power.. not really!... it was to "please" OEMs that have all major components on "mobile" platforms certified for PCIe.

The question is if you have to go for PCIe, then put that crap wasting precious SHP space controller on-die, that will save some power of course, compared with putting the crap on the chipset made of older and less efficient processes(Ed)...

Time to market!.. AMD could had gone for HT links, but then it would had to design an equivalent "desktop" chipset and certify it for mobile.. it would had been bigger and hotter and more time to develop, specially the north-bridge chipset HUB, because those chipsets are still made on 65nm "bulk" instead of 32nm gate-first HKMG, compared with a chipset HUB which has the major space wasting PCIe controller on the CPU/APU die (Ed) ..

Mobile is considerably slower in operation.. for Mobile, HT links make little difference, and the big question is configurations with an IGP or a discrete GPU on the same platform. AMD is delivering those discrete GPUs on PCIe interface.. IF and when those interfaces change to HT, then it will be a complete different story...

So to trump Intel on mobile, the whole strategy is hampered by a crap little thing(PCIe) that so few ppl pay attention to... why is that ?... because if those interfaces of discrete GPU offerings is HT, the best Notebooks would have to have AMD all around (INTEL WILL NEVER TOUCH "hypertransport" BY PROPAGANDA AND CONTROL REASONS(its out of their control).. EVEN IF ITS CLEARLY SUPERIOR)..

.. that is why Intel rushed so quickly SB out, and lucky because the "bug of rush" was on the chipset not the CPU itself .. that is why GPGPU compute is not exploding (GPGPU on HT makes the most sense), its hold back... that is why "here" we get "shills" talking about "CPU acceleration" :lol: ... that is why the agreement with Nvidia, if Intel is blocked in AMD side because of GPGPUs on HT interface, it will go and "spin in-grand" their mobile offerings with Nvidia GPUs instead...
Does GPGPU have a bright future? - Blog - StreamComputing
OpenCL on the CPU: AVX and SSE - Blog - StreamComputing

Transfer times(latency) play a very important roll in OpenCL.. that is why there is OpenCL on CPUs, and that is why APUs is a winning design.. AND THAT IS WHY HyperTransport MAKES A HELL LOT OF SENSE!( HTX also)

Waiting for Mobile OpenCL – Q1 2011 - Blog - StreamComputing
SSEx, AVX, FMA and other extensions through OpenCL - Blog - StreamComputing

disruptive indeed

disruptive-technologies.pdf (Objecto application/pdf)

AMD between a rock and a hard place!.. that is the whole question about PCIe vs HT links!... its strategy, not any major technical hurdles. Until AMD frees itself from the straitjacket, it will never be really able to challenge Intel on this crucial segment of mobile.

Zacate was a good development.. it could serve not only for ultramobile/embedded which is aimed at, but it could go much higher above in the food chain... Had it be on an HT link chipset and all things would had been solved.

But OEM pressure demanded PCIe compliance, so that " the spin market" could go for ITX with PCIe slots and then "usual suspect" sites spin comparatives with CLEAR Intel desktop offerings downgraded to the same, to dismay those superior ultramobile AMD offerings.

The solution might be for AMD to push hard GPGPU computing and make mobile chipsets out of 28nm gate-first HKMG "bulk" processes. That would make them small enough(up to 4x smaller than a 65nm similar), power efficient enough and cheap enough... and make the BEST (now its GPGPU with hardware schedulers to) "mobile" discrete GPGPU offerings on HT links FIRST!!!(not discarding PCIe entirely, but the best go for HT interface first, for not helping sell Intel CPUs instead of AMD CPU/APUs)..

On the CPU/APU(CPU+GPU in Intel, not really fusion); side its hard for AMD to establish a superior product, counting the all "rigged tests" and spin that goes around!.. on the GP/GPU side it might be much easier!.. that is why Intel gave Nvidia 1.5B $, more than the 1.2B $ it gave to AMD for settlement(peanuts for them!), for the multi-billions it stole from AMD due to their world wide condemned business practices.

So its not technical.. and in my POV all resolves around this issue of interconnect. I've been ranting about it for years!.. i might stop posting for 2 years, then came back, but if AMD hasn't the "guts" to free itself from the crap, when i re-start posting it would be the same s**t allover...

A good example its that "slide" of "future APU" features.. why wait for ccPCIe, that will be IF only Intel pleases... and when Intel pleases!... so it might go far far into the future than those planned 4th Phase features time schedule => AMD has no control over it at all!... when with "cache coherent Hypertransport", AMD could put the feature even in the 3th Phase time schedule!.. its really mind boggling !!!?? :shock: :lol: :roll: :roll:

Does AMD board "HATE" hypertransport !??.. is there an Intel "YURI" sitting down!??... i can't understand!!??... its an easy exercise i think!.. at least me as AMD's CEO, and i can imagine myself sending chocolate boxes to Otellini with nice notes!( he could put them next to the First Billion plaque of stolen shipments!)... Is that the reason why the turmoil at AMD around the CEO place !??... an "yuri" un-catched and its the end of AMD in a long term...

There are rumors.. example...
Kitguru says intel is pulling in roadmap thanks to BD...

But in my POV, it might not exactly have to do with "Orochi" itself.. alarm bells ringed at Intel when they realized that AMD is not exactly following the "dictatum".. no PCIe links on-die for that!?.. might indicate a "revolution", and a superior Trinity without PCIe as major interconnect, followed by GPGPUs on HT interfaces also.. and Intel might be for a good deal of trouble...( EVEN on MOBILE a direct APU < cache coherent hypertransport > discrete GPGPU, its a whole lot similar to some server/wokstation(HPC) configs of "torrenza"!... and OpenCL makes the whole transparent to developers... so why not scale that all around even to clusters?)

Its my impression or Intel has been quiet but now its intensifying the "spin" again.. at least a comparative of a desktop offering with Zacate, its very stupid if anyone ask me!
Last edited by mmarq on Tue Apr 19, 2011 8:32 pm, edited 2 times in total.
mmarq
K8 Athlon 64 (Orleans) Expert Boarder
K8 Athlon 64 (Orleans) Expert Boarder
 
Posts: 2337
Joined: Sat Jul 14, 2007 4:31 am

Re: Charlie says that 28nm Bobcate follow up has taped out..

Postby mmarq » Tue Apr 19, 2011 8:17 pm

AussieFX wrote:I think Llano had all it's outside ht links removed to save power/space so it's out but BDv2 could certainly fulfil that role.


Following above.. yes it might. BD v2 its my guess is mostly a macro-architecture improvememt with "intelligent caches" and a sophisticated memory controller... the uarch should maintain basically unchanged... and no need yet to go for embedded DRAM in a socket with interposers like S|A Charlie is advancing for IvyBridge...

And the best aspect of it is that those advanced cache systems and IMC, has much more IPC improvement potential than DRAM on interposers..

So taking advantage that the effort is on macro-arch and links, and pimping the ride for GPGPUs on HT links would be quite an achievement, quite ahead of any possible "interposed" IvyBridge.. if anyone asks me..

But we have to wait to see what the "yuri" can destroy or block from the inside, because from the outside if AMD doesn't push it hard(subsidizing it if must), the reception is going to be cold... unless the public wise up and demand for it!(=> much less morons around and Intel is doomed.. or at least in serious trouble!..)
mmarq
K8 Athlon 64 (Orleans) Expert Boarder
K8 Athlon 64 (Orleans) Expert Boarder
 
Posts: 2337
Joined: Sat Jul 14, 2007 4:31 am

Re: Charlie says that 28nm Bobcate follow up has taped out..

Postby BaronMatrix » Mon Apr 25, 2011 6:56 pm

All very interesting comments. I agree that it's prudent to shrink the chipsets now. GF can install 45nm equipment from Dresden in Chartered and use that. 65nm to 45nm is a helluva die-savings. They should definitely split the load for GPUs and get some SHP bulk experience at GF. But then, since they have an SOI GPU now why not do more. I'm sure that BD will allow for even more SPs on the die since it'll be much more power efficient.
cPCIe does have the advantage of not leaving out the nVidia option, especially since 990FX and beyond will have SLI support. Plus there are PCIe accelerators out now. I was really surprised when nVidia didn't throw money at AMDs platform after Intel canceled their chipsets. They could have made more of a market with a little optimization for PhenomII. It is still neck and neck with Intel in games by price.

Also, Seifert mentioned "a couple of 28nm tapeouts" in the transcript for the call at seekingalpha. I've been hearing that TSMCs 28nm isn't ready and GFs was ready in Feb, I believe, so they must have been at GF.
MSI 790GX X4 940 - @ 3GHz - AC64 HSF - HD 4870 512MB - 8GB DDR2 - 1TB HDD - BluRay - TX650 W - Antec 300 - 2X ACER 23"
HP DV7-3065 - M600 - 17" LCD - BD-Rom - HDMI\eSATA - .5TB HDD - HD4200
User avatar
BaronMatrix
K8 Athlon 64 (San Diego) Expert Boarder
K8 Athlon 64 (San Diego) Expert Boarder
 
Posts: 1949
Joined: Mon Jan 29, 2007 3:13 am

Re: Charlie says that 28nm Bobcate follow up has taped out..

Postby Smartidiot89 » Tue Apr 26, 2011 9:16 am

mmarq wrote:
Reiuji Utsuho wrote:Ordinary Kill-a-watt measurement makes no sense in low power platforms. Most of the power is lost during conversion. Linus tech tips measured 32W for a C-50 tablet, but that thing runs 6 hours on a 28Whr battery.


Reiuji Utsuho wrote:Low TDP does not make E-350 a power efficient processor, just because its raw performance is too weak. It's not surprising E-350 consumes more power than i3 2100T in some of the tests:


Where do you guys came from ??.. its obvious that i3 has more performance than a zacate, maybe its performance its "weak", but no where is i3 double the performance of Zacate(even on the biased xbitlab test).. yet it might consume almost double the power, as a "WHOLE" system.. so you talk about performance/watt or performance/$ !?... (Intel loses in either case!)

Your conclusions make me :lol: ... when so obvious contradictions doesn't make you think.. and pause ??

I have the "Washington Monument" to sell if you are interested "Reiuji Utsuho" !?

From your own post isn't obvious that a whole Zacate system consumes less than a single Intel CPU.. which BTW, as seen before, might very well have power management tricks making those measurements dubious and at a low level standard, because of optimized benchmarketoses even the ones for measuring power, and so "avoids" NORMAL SPIKES that might put it way above in average power consumption !??...

Show me "measurements at the wall".. on tests that might consider multitasking... and don't be surprised(at least i wont) to see that a single Intel i3 CPU SYSTEM consumes almost the DOUBLE of a whole Zacate SYSTEM ...

[EDIT: i might even consider low end desktop... if you put a whole lot of "low end" in it.. but Zacate its clearly for ultramobile/embedded systems.. which SB is not(CLEARLY).. so again the problem is ORANGES vs Apples.. and so this thread is derailed about an obvious biased test from a "site" that has made ppl at this site in the past, scream at so many obvious "twists" ]

Total power consumption to get X task done is what Reiuji Utsuho. If we look at true powerconsumption and not the joke that TDP is I am sure Zacate will look even more powerefficient on paper, but if it for example consumes 50% less and takes 150% longer to finish the task the one which doesn't look that powerefficient on paper suddenly is.

Also in the server-space sometimes you're better off buying a high-TDP CPU if it will be under heavy load all the time then one off those low-TDP models, cause the former will in many cases use less watt per "task".

I love Zacate by the way, and is digging around my wallet considering if it's something I can afford during this summer to make a HTPC :P
Smartidiot89
 

Re: Charlie says that 28nm Bobcate follow up has taped out..

Postby MKruer » Tue Apr 26, 2011 5:57 pm

think there needs to be a distinction between time to complete a task and power efficiency over time.

You might be correct about the 50% less power but 150% longer to process. However when you measure system over the same amount of time is it really more efficient?

The Zacate is targeted at the mobile end were so it probably targeted at longevity. vs short powerful bursts.

I an not saying that either is wrong but it worth take a look at a different point of view
Lian-Li PC-V2000 Plus Aluminum Case; Seasonic S12 Energy+ 550 PSU; Asus M4A785TD-V EVO; Phenom II X4 965 Black Edition C3 @ 4.0Ghz ; Thermalright Ultra-120 eXtreme Rev.C; 8GB OCZ AMD Black Edition @ 1333Mhz; Sapphire Radeon HD 7870
User avatar
MKruer
K10 Opteron (Barcelona) Administrator
K10 Opteron (Barcelona) Administrator
 
Posts: 2439
Joined: Mon Mar 01, 2004 4:21 am
Location: I am not paid to do this, I don't even like to do this, I wonder why am I still doing this?

Previous

Return to AMD Fusion, Bobcat, Bulldozer

Who is online

Users browsing this forum: No registered users and 1 guest