TechReport on Barcelona

Printable View

Show 100 post(s) from this thread on one page

09-12-2007, 11:53 AM
informal

And the whole "testing" was done in a hurry and not in detail(plus some tests favored intel since they were compiled with highly optimized Intel binaries)
09-12-2007, 11:56 AM
Hornet331

Quote:

Originally Posted by gallag

I cant understand how people believe AMD sent out bugged systems for review. I would near blindly accept this if they obtained them from a third party but AMD sent them, Seriously, Do you think this is likely?

Also it is pretty funny that a lot of amd fans are saying "I am not disappointed, k10 is great, I am very very happy with it,BUT ALSO DONT BELIEVE ANY OF THE REVIEWS i AM BASING THIS GREATNESS ON BECAUSE THEY HAVE BEEN DONE WITH BUGGED CHIPS AND MAKE IT LOOK BAD" LOL

What you see is what you get.

its easier to live with a lie, than live with the truth. :p:

anyways ill wait till november, till than solid numbers should have turned up.
09-12-2007, 12:02 PM
Periander6

Quote:

Originally Posted by gallag

I cant understand how people believe AMD sent out bugged systems for review. I would near blindly accept this if they obtained them from a third party but AMD sent them, Seriously, Do you think this is likely?

Also it is pretty funny that a lot of amd fans are saying "I am not disappointed, k10 is great, I am very very happy with it,BUT ALSO DONT BELIEVE ANY OF THE REVIEWS i AM BASING THIS GREATNESS ON BECAUSE THEY HAVE BEEN DONE WITH BUGGED CHIPS AND MAKE IT LOOK BAD" LOL

What you see is what you get.

Not only are they willing to believe that AMD is a bunch of brain-damaged clowns for sending out bugged and crippled chips and systems directly to reviewers only 3 days before launch, but also that they are so bloody brain-damaged that they don't even say anything about it in the days afterwards.

YOU WOULD THINK THAT BY NOW AMD WOULD HAVE GOTTEN AROUND TO SAYING "OH HEY, WE WUZ JUST JOKING SENDING YOU THOSE CRIPPLED CHIPS FOR YOU TO REVIEW FOR ALL THE WORLD, HERE ARE THE REAL PARTS NOW"

But no, the story doesn't have to make a bit of sense as long as it keeps the hope and hype alive and hey some guy on a message board said it was so. That's credible isn't it?

:rofl:
09-12-2007, 12:07 PM
DoubleZero

Quote:

Originally Posted by savantu

Oh yes...did it ever occurred to you that AMD's rise was due to ATI's acquisition ?

lol, and your point is?
I guess AMD bougth ATI for something... :shrug:

Funny that when it comes to revenue ATI is relevant, but when it comes to debt it seems AMD never bought ATI.

Quote:

Originally Posted by savantu

Secondly , would you bet for AMD staying in top 10 this year ? ;)

Well yes since AMD reached the Top10 THIS YEAR, with a 4 year old product, and now has a new one.

http://arstechnica.com/news.ars/post...p-10-list.html
;)
09-12-2007, 12:15 PM
gallag

Quote:

Originally Posted by Carfax

They had to meet the launch date.

For God's sake, did you even read the reviews? In some benches, there was practically NO IMPROVEMENT over the K8, and in certain synthetic benches ie memory latency, the performance was worse compared to the K8.

These chips were bugged, thats a fact. The reviews were not done on retail samples.

mabey using l3 is making latencys worse?

Quote:

Originally Posted by Periander6

Not only are they willing to believe that AMD is a bunch of brain-damaged clowns for sending out bugged and crippled chips and systems directly to reviewers only 3 days before launch, but also that they are so bloody brain-damaged that they don't even say anything about it in the days afterwards.

YOU WOULD THINK THAT BY NOW AMD WOULD HAVE GOTTEN AROUND TO SAYING "OH HEY, WE WUZ JUST JOKING SENDING YOU THOSE CRIPPLED CHIPS FOR YOU TO REVIEW FOR ALL THE WORLD, HERE ARE THE REAL PARTS NOW"

But no, the story doesn't have to make a bit of sense as long as it keeps the hope and hype alive and hey some guy on a message board said it was so. That's credible isn't it?

:rofl:

:rofl:
09-12-2007, 12:27 PM
savantu

Quote:

Originally Posted by DoubleZero

lol, and your point is?
I guess AMD bougth ATI for something... :shrug:

Funny that when it comes to revenue ATI is relevant, but when it comes to debt it seems AMD never bought ATI.

Well yes since AMD reached the Top10 THIS YEAR, with a 4 year old product, and now has a new one.

http://arstechnica.com/news.ars/post...p-10-list.html
;)

Umh , that's for 2006.I'm asking about 2007.
09-12-2007, 12:31 PM
Carfax

Quote:

Originally Posted by savantu

:ROTF:

Still playing the old disk ?

The systems were sent by AMD , secondly IBM published benchmarks too and they used a B1 chip , Anandtech used B2.So AMD screwed IBM and the public ? Mind you , AMD's own benchmarks were done on B1 or earlier.

Bx chips do not have performance bugs , they have only different scaling capabilities.End of story.

The B2 chips still has the memory controller bugs.

According to Graham, the BA stepping solves the memory performance issues.

Just look at the review! You can tell something is wrong with the memory performance of these chips.

The latency benches that Tech-Report conducted were abismal.

The SSE optimized benches also weren't up to par, considering the doubled SSE throughput.

If this is the final performance we can expect from the K10, then AMD screwed up horribly.
09-12-2007, 12:58 PM
DoubleZero

Quote:

Originally Posted by savantu

Umh , that's for 2006.I'm asking about 2007.

Nevermind it was retarded isupply saying AMD was top10 in 2006 when it was not... i thought they were saying 1H 2007, it was the only thing that made sense.

Real numbers:
http://www.electronics.ca/presscente...006/Page1.html
http://www.electronics.ca/presscente...ing/Page1.html

Both AMD and Intel raised from 1H 2006.
09-12-2007, 01:04 PM
accord99

Quote:

Originally Posted by informal

In case you missed it,look at what freq. intel quad must work to attain the same level of performance(i even bolded it for you).

About 1.9GHz with Techreport's review.
09-12-2007, 01:12 PM
gallag

Look, why don't people just accept what it is, K10 is nice, it has near the same ipc as conroe and is very thrifty with power usage, Its a nice chip but the main people hurting AMD at the minute are all the AMD fanboys hypeing the hell out of it and always raising expectations not only that but do you not think you are hurting the company you love so much by insisting that they have launched buggy chips, do you not want them to sell these bloody things?

Just accept it and look forward to higher clocks on phenom
09-12-2007, 01:30 PM
gallag

Quote:

Originally Posted by FLG_Poncho

Ummm.... correct me if I'm wrong, but isn't the range of TDP for AMD's quad core from 80-120 watts, 120 being for the higher clocked parts that are not even out yet? While I agree... 80 watts is good for a quad, it certainly doesn't blow my skirt up. :D Come on now... Intel has 50watt quad cores at 2.13. Keep in mind that these are server CPUs therefore are binned, where as the desktop ones are not. I'm sure this will be the same case with AMD when they launch their desktop line.

At the same time though Intel is stuck with fb dimms and looking at the reviews so far amd has better performance per watt with k10 (at least till 45nm:) )
09-12-2007, 02:04 PM
Hornet331

Quote:

Originally Posted by FLG_Poncho

I agree... FBDIMMs kinda kill the power argument, well... that and the southbridge, northbridge. :D But this is a CPU discussion. :D

Oh, and BTW with San Clemente you'll lose all the craptastic power adders to Intel's platform. ICH9 + DDR2 FTW!!! :D

dual cpu platform with unreg ddr2? nice! :yepp:
09-12-2007, 04:21 PM
Hornet331

wow thats sure nice to hear. :)
09-12-2007, 05:15 PM
alfaunits

Quote:

Originally Posted by FLG_Poncho

I'm 90% certain there will be a UDIMM (non registered) version... 100% sure there is a RDIMM only version.

When and how much!? :-)
09-12-2007, 09:39 PM
savantu

Quote:

Originally Posted by alfaunits

When and how much!? :-)

November.

Cheap DP segment.
09-12-2007, 09:56 PM
savantu

Quote:

Originally Posted by Carfax

The B2 chips still has the memory controller bugs.

According to Graham, the BA stepping solves the memory performance issues.

Just look at the review! You can tell something is wrong with the memory performance of these chips.

What ? From what I see they are better than the K8 with the same memory.

http://techreport.com/r.x/opteron-23...-cache-1gb.gif

Quote:

The latency benches that Tech-Report conducted were abismal.

Well , my dear Watson , there is an L3 on chip now which can add up to 38ns of latency.Did you expect no trade offs whatsoever ?

Quote:

The SSE optimized benches also weren't up to par, considering the doubled SSE throughput.

I think you need to visit an eye specialist.

~3.7x with 2x the cores
http://techreport.com/r.x/opteron-23...dra-mm-int.gif

~3x with 2x the cores
http://techreport.com/r.x/opteron-2300/sandra-mm-fp.gif

Where does it say that K10s SSE units must have the same throughoutput as Core's ?

IIRC , K10 has 2 SSE units and Core has 3.In other words Core has better INT/FPU power and it shows .

K8/10 show better FPU power than Core sometimes , not because they have more resources , but due to better memory BW+latency.FP apps like huge memory BW.

Quote:

If this is the final performance we can expect from the K10, then AMD screwed up horribly.

Don't tell me you believed there's some magic dust inside ?
09-13-2007, 01:58 PM
Richie P

Quote:

Originally Posted by Richie P

I may be wrong, please correct me if i am :)

Quote:

Originally Posted by hollo

from what i've read the l3 cache is a spilloff cache, and the memory controller feeds the l2, so the CPU isn't held up by the l3 latency (which would result in a pretty bad performance hit). cache latency (seconds) = latency (clocks) * duration of a clock (seconds)

and duration of a clock (seconds) = 1/frequency (to get seconds per cycle instead of cycles per second)

so latency (clocks) = latency (seconds) / (1/frequency) = latency (seconds) * frequency

using 23ns @ 2ghz, 19ns @ 2.5ghz:
(23*10^-9)*(2*10^9) = 46 clocks
(19*10^-9)*(2.5*10^9) = 47.5 clocks

latency in clockcycles increased going to 2.5ghz, there's no way you'll get better than linear scaling... that was a myth

Sorry to dig a mildly old thread, but I wanted to thank hollo for teaching me in such a nice and informative way. I didn't claim to know everything and wanted to be corrected if there was something I missed, so thank you :up:
09-13-2007, 02:14 PM
Shintai

Just because AMD aint super crown of the CPUs doesnt mean they are screwed. Try ask Intel how many 3Ghz quads they sell. Its not alot.

What you need is a product you can make with competitive price/performance in the high volume segments.

Ofcource K10 aint looking good on that due to its monolithic diesize. But atleast AMD got the hint for the next time and going the MCM way with all its designs.

K10 now needs better yields and faster speed. There is no magic steppings, broken SSE/FP or whatever. Its also a demonstration on there is a difference between a 1-1.5B$ R&D budget and a 5.5-6B$ R&D.

Only 2 things to blame, native design and the now doubleedged sword of SOI that strikes back.

Had AMD started with MCM K8 quads and then later on MCM K10s from the start they would be in an alot better situation on that alone. 150$ 2X2 brisbanes with a higher margin that would sell like ice melts in mexico. Along with staying with the bulk design and forget IBMs more or less exotic approaches. Just like Sony got caught in the Cell nightmare. Then AMD would simply sit with better cards today.

In short, fire Hector :p:

Oh, and left ATI to die slowly as they were. What a waste of cash. Else they could have picked ATI up today for 1/3th of the price.
09-13-2007, 02:51 PM
doompc

Quote:

using 23ns @ 2ghz, 19ns @ 2.5ghz:
(23*10^-9)*(2*10^9) = 46 clocks
(19*10^-9)*(2.5*10^9) = 47.5 clocks

http://www.hardwaresecrets.com/imageview.php?image=7812

NB clock doesn't increase equal to CPU clock.
L3 cache latency must increase @ higher CPU Clock.
09-13-2007, 03:10 PM
xlink

i see later revisions offering 5-20% greater performance per clock and scaling to 3Ghz. let's see if AMD can have that within 6months ... because that's when I'm upgrading...
09-13-2007, 03:17 PM
Zytek_Fan

Sign me up for 2 of the 2.4ghz ones.
09-13-2007, 03:23 PM
Hornet331

Quote:

Originally Posted by xlink

i see later revisions offering 5-20% greater performance per clock and scaling to 3Ghz. let's see if AMD can have that within 6months ... because that's when I'm upgrading...

20% higher ipc.... :shocked:

even intel only claimed an avarage 10% ipc increase with penryn (which is in reallity between 5-<10%) and that thing is by far more then a core revision...

your quite optemistic. :yepp:

i can see higher clocks and the lower voltages and maybe a moderate ipc advantage in the 1-3% range, but 20%... wow.

Show 100 post(s) from this thread on one page