Tech Report Woodcrest preview - Woodcrest 3.0 VS. Opteron 2.6

[PetNorth]

here:

http://www.techreport.com/onearticle.x/10021

Considering that top of the line 3.0ghz Woodcrest is "only" 15-20% overall faster than DC Opteron 285 2.6ghz, isn't impressive at all if you ask me, specially with 15% more clock speed than Opty :disgust:

 

[robertk2012]

Thats not even close to being a comparison for desktop usage. The intel is even using onboard graphics while the amd has a PCI-E card.

 

[SexyK]

Also, the Woodcrest machine is using FB-DIMMs which are great for when you need large amounts of memory, but don't offer the same performance as vanilla DDR2. Doesn't seem to be the best apples-to-apples comparison.

 

[PetNorth]

well they use common desktop apps, where 1. graphic card is absolutely irrelevant and 2. I guess memory type is what Woodcrest uses and btw with these benches (rendering, audio encoding, video encoding) memory is basically irrelevant too.

 

[SexyK]

Woodcrest supports both registered/ecc DDR2 and FB-DIMM configurations. What performance impact the platform choice will have remains to be seen however

 

[zephyrprime]

Look at this picture:
core pictures

OMG the Tulsa die is humongous! It's almost as big as an itanium.

 

[kknd1967]

well I see 5-40%, many are around 30%
and sometimes 400%

One can sure overclock an Opteron to 3G. But the performance most likely will not scale linearly and then you need to consider the TDP increase.

btw, has anyone seen the same chip, performance can scale faster than linear in clock freq? Just curious.

Originally posted by: PetNorth
here:

http://www.techreport.com/onearticle.x/10021

Considering that top of the line 3.0ghz Woodcrest is "only" 15-20% overall faster than DC Opteron 285 2.6ghz, isn't impressive at all if you ask me, specially with 15% more clock speed than Opty :disgust:

 

[robertk2012]

Originally posted by: PetNorth
well they use common desktop apps, where 1. graphic card is absolutely irrelevant and 2. I guess memory type is what Woodcrest uses and btw with these benches (rendering, audio encoding, video encoding) memory is basically irrelevant too.

1. graphic card is revelavent since it pulls system memory and effects all around performance. Memory is not irrelevant especially with an intel platform.

2. AMD was soundly beaten in every test. How often does that happen. Even when AMD was dominate Intel would get a few wins.

 

[Eug]

It's faster than the Opteron and lower power. What's not to like?

Or were you expecting something completely outrageous, like twice the speed or something, or 40 Watt TDP at 3 GHz?

P.S. Already posted here.

 

[PetNorth]

Considering Intel claims 20-40% faster at same clock speed, I can't see it...

Cinebench overall Woodcrest 3.0ghz 15% faster than Opteron 2.6ghz
POV-Ray 4 threads Woodcrest 3.0ghz 6% faster than Opteron 2.6ghz
3DS Max overall Woodcrest 3.0ghz 21% faster than Opteron 2.6ghz
Windows Media Encoder Woodcrest 3.0ghz 5% faster than Opteron 2.6ghz
Xmpeg / DivX Woodcrest 3.0ghz 16% faster than Opteron 2.6ghz
Lame MT overall Woodcrest 3.0ghz 19% faster than Opteron 2.6ghz

This is about 15% overall faster (again Woodcrest 3.0ghz vs Opteron 2.6ghz)

(sintetic nonsense benches jumped)

btw 5% less power consumption at full load (it would be around 10% at same clock) isn't impressive either (specially 0.065 vs 0.090)

 

[imported_Questar]

Cinebench Render x64 multi 12.5%
Cinebench Render x64 single 17.3%
Cinebench Render x86 multi 22.2%
Cinebench Render x86 single 25.8%
Cinebench Shading x64 15.2%
Cinebench Shading x86 11.9%
3DMax Render 22%
3DMax Metalray 21.5%
Windows Media Encoder 5.5%
DivX6 15.9%
Lame CBR MS single 27.7%
Lame CBR MS multi 26.1%
Lame CBR Intel single 28.6%
Lame CBR Intel multi 28%
Lame VBR MS single 17.2%
Lame VBR MS multi 5.2%
Lame VBR Intel single 24.1%
Lame VBR Intel multi 16.6%
Sphinx MS 23%
Sphinx Intel 24.8%
picCOLOR overall 34%

Power at idle 19%
Power at load 5%

All using 533Mhz memory instead of 667.

 

[bunnyfubbles]

Originally posted by: robertk2012

Originally posted by: PetNorth
well they use common desktop apps, where 1. graphic card is absolutely irrelevant and 2. I guess memory type is what Woodcrest uses and btw with these benches (rendering, audio encoding, video encoding) memory is basically irrelevant too.

1. graphic card is revelavent since it pulls system memory and effects all around performance. Memory is not irrelevant especially with an intel platform.

Yeah, if we were talking about NetBurst...oh wait, we're not

 

[kknd1967]

why your number is somewhat lower than I calculated?
eg. see the bold one below

Originally posted by: Questar
Cinebench Render x64 multi 12.5%
Cinebench Render x64 single 17.3%
Cinebench Render x86 multi 22.2%
Cinebench Render x86 single 25.8%
Cinebench Shading x64 15.2%
Cinebench Shading x86 11.9%
3DMax Render 22%
3DMax Metalray 21.5%
Windows Media Encoder 5.5%
DivX6 15.9%
Lame CBR MS single 27.7%
Lame CBR MS multi 26.1%
Lame CBR Intel single 28.6%
Lame CBR Intel multi 28%
Lame VBR MS single 17.2% 20.8%
Lame VBR MS multi 5.2% 5.5%
Lame VBR Intel single 24.1% 31.8%
Lame VBR Intel multi 16.6% 20%
Sphinx MS 23% 29.9%
Sphinx Intel 24.8% 32.9%
picCOLOR overall 34%

Power at idle 19%
Power at load 5%

All using 533Mhz memory instead of 667.

 

[Markfw]

Originally posted by: Eug
It's faster than the Opteron and lower power. What's not to like?

Or were you expecting something completely outrageous, like twice the speed or something, or 40 Watt TDP at 3 GHz?

P.S. Already posted here.

I think he means that by the time you can buy one, there may be a 65 nm 3 ghz Opteron that equals or beats it, and at a lower power yet.

 

[TuxDave]

Originally posted by: Markfw900

Originally posted by: Eug
It's faster than the Opteron and lower power. What's not to like?

Or were you expecting something completely outrageous, like twice the speed or something, or 40 Watt TDP at 3 GHz?

P.S. Already posted here.

I think he means that by the time you can buy one, there may be a 65 nm 3 ghz Opteron that equals or beats it, and at a lower power yet.

I don't think they have 65nm Opterons on the roadmap for June.

 

[imported_Questar]

Originally posted by: Markfw900

Originally posted by: Eug
It's faster than the Opteron and lower power. What's not to like?

Or were you expecting something completely outrageous, like twice the speed or something, or 40 Watt TDP at 3 GHz?

P.S. Already posted here.

I think he means that by the time you can buy one, there may be a 65 nm 3 ghz Opteron that equals or beats it, and at a lower power yet.

What's TDP on a FX-62? 125 watts?
Good luck to AMD gettig that down 50%.

 

[imported_Questar]

Originally posted by: kknd1967
why your number is somewhat lower than I calculated?

Beats me, how did you calculate them?

 

[robertk2012]

Originally posted by: bunnyfubbles

Originally posted by: robertk2012

Originally posted by: PetNorth
well they use common desktop apps, where 1. graphic card is absolutely irrelevant and 2. I guess memory type is what Woodcrest uses and btw with these benches (rendering, audio encoding, video encoding) memory is basically irrelevant too.

1. graphic card is revelavent since it pulls system memory and effects all around performance. Memory is not irrelevant especially with an intel platform.

Yeah, if we were talking about NetBurst...oh wait, we're not

Ok so your telling me that slow ddr2 533 memory is not going to have an effect. Remmeber he is trying to say conroe will suck because of these benches. Intel still needs the extra bandwith and it may not be to extent it once was but it still has a definate impact on performance.

Also conroe is coming soon and dont see a 3 ghz opty out there as we dont even have a 2.8 ghz opty or is there even one being tested.

 

[robertk2012]

Originally posted by: TuxDave

Originally posted by: Markfw900

Originally posted by: Eug
It's faster than the Opteron and lower power. What's not to like?

Or were you expecting something completely outrageous, like twice the speed or something, or 40 Watt TDP at 3 GHz?

P.S. Already posted here.

I think he means that by the time you can buy one, there may be a 65 nm 3 ghz Opteron that equals or beats it, and at a lower power yet.

I don't think they have 65nm Opterons on the roadmap for June.

Sure they do. June 2007 when intel has 45 nm 4 core processors. They already have working versions running and may be up to 8 cores by the end of 2007.

If I was AMD I would skip 65 and start working on 45 if not they are going to be so far behind they may not catch back up.......that is until intel starts taking their success for granted again.

Also these are dual chip systems so again I state that these benchmarks have no bearing on conroes performance.

 

[kknd1967]

LAME VBR Single
We are talking about speed, so speed = workload/time
speed woodcrest = workload/22
speed opteron = workload/29

woodcrest/opteron = (w/22) / (w/29) = 29/22=1.3182

now I realize maybe you are doing 1-22/29, which is not really a performance ratio

Originally posted by: Questar

Originally posted by: kknd1967
why your number is somewhat lower than I calculated?

Beats me, how did you calculate them?

 

[imported_Questar]

Originally posted by: kknd1967
LAME VBR Single
We are talking about speed, so speed = workload/time
speed woodcrest = workload/22
speed opteron = workload/29

woodcrest/opteron = (w/22) / (w/29) = 29/22=1.3182

now I realize maybe you are doing 1-22/29, which is not really a performance ratio

Originally posted by: Questar

Originally posted by: kknd1967
why your number is somewhat lower than I calculated?

Beats me, how did you calculate them?

29-22 = 7
7/29 = 24.1%

 

[kknd1967]

no offense.

but I think your measurement is not related to performance
A finishes in 1 second
B finishes in 10 seconds
A is 10 timers faster than B, instead of 9/10=90% faster, right?

Originally posted by: Questar

Originally posted by: kknd1967
LAME VBR Single
We are talking about speed, so speed = workload/time
speed woodcrest = workload/22
speed opteron = workload/29

woodcrest/opteron = (w/22) / (w/29) = 29/22=1.3182

now I realize maybe you are doing 1-22/29, which is not really a performance ratio

Originally posted by: Questar

Originally posted by: kknd1967
why your number is somewhat lower than I calculated?

Beats me, how did you calculate them?

29-22 = 7
7/29 = 24.1%

 

[Eug]

Originally posted by: PetNorth
Considering Intel claims 20-40% faster at same clock speed, I can't see it...

Cinebench overall Woodcrest 3.0ghz 15% faster than Opteron 2.6ghz
POV-Ray 4 threads Woodcrest 3.0ghz 6% faster than Opteron 2.6ghz
3DS Max overall Woodcrest 3.0ghz 21% faster than Opteron 2.6ghz
Windows Media Encoder Woodcrest 3.0ghz 5% faster than Opteron 2.6ghz
Xmpeg / DivX Woodcrest 3.0ghz 16% faster than Opteron 2.6ghz
Lame MT overall Woodcrest 3.0ghz 19% faster than Opteron 2.6ghz

This is about 15% overall faster (again Woodcrest 3.0ghz vs Opteron 2.6ghz)

(sintetic nonsense benches jumped)

btw 5% less power consumption at full load (it would be around 10% at same clock) isn't impressive either (specially 0.065 vs 0.090)

1) I always take the claims by the manufacturer with a teaspoon of salt.
2) 15% faster and 5% lower power is still significant. Even if AMD by July could up the performance by 10% and keep the power usage the same as now, Intel will still have the overall lead. I think that's quite an impressive turnaround for Intel, considering the comparatively poor performance/Watt ratio in their previous server chips.

 

[PetNorth]

Originally posted by: robertk2012
Remmeber he is trying to say conroe will suck because of these benches.

I don't remember I'm trying nothing. Thanks.

we dont even have a 2.8 ghz opty or is there even one being tested.

early june.

 

[PetNorth]

Originally posted by: Eug

Originally posted by: PetNorth
Considering Intel claims 20-40% faster at same clock speed, I can't see it...

Cinebench overall Woodcrest 3.0ghz 15% faster than Opteron 2.6ghz
POV-Ray 4 threads Woodcrest 3.0ghz 6% faster than Opteron 2.6ghz
3DS Max overall Woodcrest 3.0ghz 21% faster than Opteron 2.6ghz
Windows Media Encoder Woodcrest 3.0ghz 5% faster than Opteron 2.6ghz
Xmpeg / DivX Woodcrest 3.0ghz 16% faster than Opteron 2.6ghz
Lame MT overall Woodcrest 3.0ghz 19% faster than Opteron 2.6ghz

This is about 15% overall faster (again Woodcrest 3.0ghz vs Opteron 2.6ghz)

(sintetic nonsense benches jumped)

btw 5% less power consumption at full load (it would be around 10% at same clock) isn't impressive either (specially 0.065 vs 0.090)

1) I always take the claims by the manufacturer with a teaspoon of salt.
2) 15% faster and 5% lower power is still significant. Even of AMD can up the performance by 10% and keep the power usage the same as now by July, Intel will still have the overall lead. I think that's quite an impressive turnaround for Intel, considering the comparatively poor performance/Watt ratio in their previous server chips.

again, considering Intel claims 20-40% faster at same clock speed.

What part you don't understand?

edit: And not only claims, but Intel IDF benches (remember?)