Intel Pentium D 920 and Pentium D 930 review

[Gamer X]

Link

 

[Duvie]

http://www.xbitlabs.com/articles/cpu/display/pentium-d-920-930_2.html

Look at the power consumption...If that doesb't say it all or prove that AMD TDP numbers and INtels are not figured the same and apples and oranges.....Like I said!!!

 

[Duvie]

http://www.xbitlabs.com/articles/cpu/display/pentium-d-920-930_9.html
http://www.xbitlabs.com/articles/cpu/display/pentium-d-920-930_10.html

So what does this tell us??? That a 2.6ghz dual core is better then a presler at 4ghz???? Pretty much doesn't look like much has actually changed since the smithfields...in video and audio that added cache has little effects.....A 3800+ X2 can hit near FX60 levels and above...get opteron 165 and I have seen most hit FX60 with the cache (cache not a factor) on air and beat or at least equal both of the preslers according to these test....


"Conclusion

The youngest Pentium D processors based on the new Presler core didn?t prove up to our expectations. Despite the improvements introduced by the manufacturer such as larger L2 cache, they are still unable to successfully compete against the rivals from the same price group in the majority of tasks. In fact, Pentium D 920 and Pentium D 930 can only prove faster than Athlon 64 X2 3800+ (the youngest dual-core processor from AMD) in office applications and selected synthetic benchmarks. That is why the only way Pentium D can become popular is due to its $60 lower price than that of Athlon 64 X2 3800+. From all other standpoint, it may not be the best purchasing decision to make at this time.

Unfortunately, the great overclocking potential allowing to speed these processors up to 4.0GHz can also hardly become a convincing argument in their favor. According to our tests, Presler based CPUs working at 4.0GHz cannot outperform Athlon 64 FX-60 working at 2.6GHz only. As a result, we have every right to claim that the overclocked Athlon 64 X2 will be mostly faster than the Presler based overclocked Pentium D processors."


No surprise even though it is much later...They could have done those test on a half as cheap NF4 board (to the Intel Asus mobo which appears one of the only board you can hit these OCs with the preslers) and the results would have been the same...



Edit: I am seeing no 60 dollars cheaper in cpu from a 920 to an 3800+...then figure in the 200 dollar Asus mobo you will need to hit 3.9-4ghz and then no contest...I would spend the extra to get an opteron and get close to 2.7-2.8ghz on air....

 

[BrownTown]

Originally posted by: Duvie
http://www.xbitlabs.com/articles/cpu/display/pentium-d-920-930_2.html

Look at the power consumption...If that doesb't say it all or prove that AMD TDP numbers and INtels are not figured the same and apples and oranges.....Like I said!!!

The TDP for these chips is 125W, the load power draw shown is 103.2W for the 930, whwere is the problem here?

 

[Duvie]

Originally posted by: BrownTown

Originally posted by: Duvie
http://www.xbitlabs.com/articles/cpu/display/pentium-d-920-930_2.html

Look at the power consumption...If that doesb't say it all or prove that AMD TDP numbers and INtels are not figured the same and apples and oranges.....Like I said!!!

The TDP for these chips is 125W, the load power draw shown is 103.2W for the 930, whwere is the problem here?

Please link...Someone said in another thread it was not that high.....If so my fault....Though AMD list like 89-95 watts for the line....Also the idle is like 1/4th

http://www.techreport.com/onearticle.x/9256

says 95 watts...smithfiled was like 130

Edit:

dude you are full of shite!!! from the same article....TDP is 95WATTS...

http://www.xbitlabs.com/articles/cpu/display/pentium-d-920-930_2.html

 

[Duvie]

95watt TDP with a amx case temp of 63.2c....Anybody want to guess they dont test at Max TCASE especially since Xbitlabs already at over 100watts...

 

[BrownTown]

k, your right, only 940 and 950 are 125. But they will all go down with the C1 stepping in April.

 

[Duvie]

Originally posted by: BrownTown
k, your right, only 940 and 950 are 125. But they will all go down with the C1 stepping in April.

except for the 920 which its states will never see c1 core stepping...Upper models are seeing c1 and highest models seeing C2....which means they are ramping up in heat as fast as the speed....Needless to say the writing is still on the wall that Conroe was the correct course for INtel to take CPU design....Dump mhz for more IPC...

So they say 95 but we see 103....AMD says like 89 and we get like 65....

One thing I should state is they are not telling us how they calculated that or how they loaded the cpus....Just to be fair...

that aside the correlation between TDP stated and actuallity now can be interpreted between the cpu listed in the article....

 

[LTC8K6]

Those heat dissipation numbers look right in the ballpark for both companies to me, given the different definitions of TDP used by Intel & AMD.

 

[dexvx]

Originally posted by: LTC8K6
Those heat dissipation numbers look right in the ballpark for both companies to me, given the different definitions of TDP used by Intel & AMD.

Do you know what TDP means? Those are powerdraw numbers, nothing to do with heat dissipation.

And obviously Xbit labs doesnt know either. They interchange "heat disspiation" with "power draw" fairly liberaly.

 

[Markfw]

Bootom line without all the crap, the 9xx series runs hotter and slower, and if OC'ing, is still more $$$$ (motherboard required for that OC)

X2's still rule for now.

 

[BrownTown]

Originally posted by: dexvx

Originally posted by: LTC8K6
Those heat dissipation numbers look right in the ballpark for both companies to me, given the different definitions of TDP used by Intel & AMD.

Do you know what TDP means? Those are powerdraw numbers, nothing to do with heat dissipation.

And obviously Xbit labs doesnt know either. They interchange "heat disspiation" with "power draw" fairly liberaly.

how exactly is the amount of power drawn not related to heat dissipation? The heat of a processor is directly related to how much power it is drawing...

 

[Zebo]

Originally posted by: Duvie
http://www.xbitlabs.com/articles/cpu/display/pentium-d-920-930_2.html

Look at the power consumption...If that doesb't say it all or prove that AMD TDP numbers and INtels are not figured the same and apples and oranges.....Like I said!!!

They are a hell of a lot better than smithfiellds which clipped 180W!!!
http://www.xbitlabs.com/articles/cpu/display/pentium4-670_3.html

Pentium 920 is 100W while 820 used 130W..
http://www.xbitlabs.com/articles/cpu/display/pentiumd-820_3.html

Thanks for posting OP - been looking for a good review.

 

[Zebo]

Originally posted by: Duvie
http://www.xbitlabs.com/articles/cpu/display/pentium-d-920-930_9.html
http://www.xbitlabs.com/articles/cpu/display/pentium-d-920-930_10.html

So what does this tell us??? That a 2.6ghz dual core is better then a presler at 4ghz???? Pretty much doesn't look like much has actually changed since the smithfields...in video and audio that added cache has little effects.....A 3800+ X2 can hit near FX60 levels and above...get opteron 165 and I have seen most hit FX60 with the cache (cache not a factor) on air and beat or at least equal both of the preslers according to these test....


"Conclusion

The youngest Pentium D processors based on the new Presler core didn?t prove up to our expectations. Despite the improvements introduced by the manufacturer such as larger L2 cache, they are still unable to successfully compete against the rivals from the same price group in the majority of tasks. In fact, Pentium D 920 and Pentium D 930 can only prove faster than Athlon 64 X2 3800+ (the youngest dual-core processor from AMD) in office applications and selected synthetic benchmarks. That is why the only way Pentium D can become popular is due to its $60 lower price than that of Athlon 64 X2 3800+. From all other standpoint, it may not be the best purchasing decision to make at this time.

Unfortunately, the great overclocking potential allowing to speed these processors up to 4.0GHz can also hardly become a convincing argument in their favor. According to our tests, Presler based CPUs working at 4.0GHz cannot outperform Athlon 64 FX-60 working at 2.6GHz only. As a result, we have every right to claim that the overclocked Athlon 64 X2 will be mostly faster than the Presler based overclocked Pentium D processors."


No surprise even though it is much later...They could have done those test on a half as cheap NF4 board (to the Intel Asus mobo which appears one of the only board you can hit these OCs with the preslers) and the results would have been the same...



Edit: I am seeing no 60 dollars cheaper in cpu from a 920 to an 3800+...then figure in the 200 dollar Asus mobo you will need to hit 3.9-4ghz and then no contest...I would spend the extra to get an opteron and get close to 2.7-2.8ghz on air....

Where's Sventy who claimed a Dually @ 3.7 beats a x2 @2.6

In reality you need 1.65 x X2 Mhz = PentD (just like my excel file predicted a year ago)

And in gaming, well forgettaboutit
http://www.xbitlabs.com/articles/cpu/display/pentium-d-920-930_11.html

 

[openwheelformula1]

3800+ X2 consistently outperforms 930 which costs more. These results are similar to gamepc's benchmarks, total domination by X2. I can honestly say netburst is simply horrible.

 

[Duvie]

Originally posted by: dexvx

Originally posted by: LTC8K6
Those heat dissipation numbers look right in the ballpark for both companies to me, given the different definitions of TDP used by Intel & AMD.

Do you know what TDP means? Those are powerdraw numbers, nothing to do with heat dissipation.

And obviously Xbit labs doesnt know either. They interchange "heat disspiation" with "power draw" fairly liberaly.

TDP = Thermal Design Power .... It is measured in watts... go read the specs at Intels own website.....

The more power it takes to run a chip at its intended frequency stably then becomes power that must be dissipated in the form of heat....The more power a cpu takes the more heat it will have to dissipate...designers of mobo, heatsinks, etc take info Intel is supplying them in terms of TDP to figure "power draws" to design thermal solution (IE Heatsink manufacturers) and power configurations on the mobo level......

Intels problem is they dont figure this on a max load (their own words) and they dont test at max Tcase temps.....

They are interrelated in every way....


http://support.intel.com/support/processors/pentium4/sb/CS-007999.htm


If not why dont you educate us??? I think you are part of the same crowd who denies Intels TDP numbers are not calculated the same as AMDs and do not represent max TDP....Eventhough I supplied you huys with the words straight from the horse's mouth...

 

[Zebo]

Intel fans dying for conroe.. hope it lives up to it's hype.

 

[Markfw]

Originally posted by: Zebo
Intel fans dying for conroe.. hope it lives up to it's hype.

I am sure they will take anything close, its like Heroin addicts looking for a fix...

Please Conroe hurry up !

 

[Zebo]

Power numbers and cross the world benchmarks may leave them short ot total domination like A64 has - but it will be most excellent - reminisint of Northwood C vs. dated Athlon XP.

 

[BassBomb]

power draw =/= power loss through heat (TDP)

 

[dug777]

Originally posted by: Zebo
Power numbers and cross the world benchmarks may leave them short ot total domination like A64 has - but it will be most excellent - reminisint of Northwood C vs. dated Athlon XP.

at the risk of sounding like the 'rollo' of the CPU forum i have a question on the whole benchmarks we've seen so far of conroe matter...while of course intel would have picked benchmarks that favoured conroe, is it reasonable to say that the results in those games/apps have traditionally been representative of overall CPU performance? Say looking at Anand's comments on the Q4...

With SMP enabled we see that Conroe holds an even larger 31% performance advantage and with it disabled, the unreleased CPU was 29% faster. If anything, Intel?s own demo was a little more conservative on Conroe and definitely not optimized to make AMD look bad.

Is it likely that the chip will be this much better in only q4 & the specific apps that intel allowed to be run (at a reasonable clock speed deficit at that), and considerably worse in other apps? Are they specifically apps that love large cache? I don't know myself, not having sat down & digested how those specific benchmarks have related to teh general overall picture (although i'm sure somone has on here & can tell me)...

I guess we will see soon enough, that's the golden rule with all this

 

[AkumaX]

Originally posted by: Duvie
http://www.xbitlabs.com/articles/cpu/display/pentium-d-920-930_2.html

Look at the power consumption...If that doesb't say it all or prove that AMD TDP numbers and INtels are not figured the same and apples and oranges.....Like I said!!!

the thing i think that's stupid is why are they comparing it to a single core a64?!???

edit: unless that's a typo, har har

otherwise, great rest of the article

mark, if conroe's cracked up what it's supposed to be, wouldn't you be on that heroin train too?

 

[Duvie]

Originally posted by: AkumaX

Originally posted by: Duvie
http://www.xbitlabs.com/articles/cpu/display/pentium-d-920-930_2.html

Look at the power consumption...If that doesb't say it all or prove that AMD TDP numbers and INtels are not figured the same and apples and oranges.....Like I said!!!

the thing i think that's stupid is why are they comparing it to a single core a64?!???

edit: unless that's a typo, har har

otherwise, great rest of the article

mark, if conroe's cracked up what it's supposed to be, wouldn't you be on that heroin train too?

i am sure they meant the X2 3800+ since the entire article compared that chip to the 920 and 930 of INtel.....

 

[LTC8K6]

Intel fans dying for conroe.. hope it lives up to it's hype.

I acknowledge that AMD chips are cooler and faster and a better value, and recommend them to people all the time, but I still run Intel chips all the time myself.

I am not waiting for Conroe at all. Certainly not dying for it.

When it comes, it will just be the latest Intel chip to me.

I never can figure out all the vitriol over who runs what.

Some folks like Hondas, some like Fords, some like Dodge, etc.
Some folks think they're all just cars....

I had plans to build AMD systems, but circumstances intervened and I just never got around to it.

AMD is better, and I run Intel. Get over it....

 

[jlbenedict]

Originally posted by: LTC8K6

Intel fans dying for conroe.. hope it lives up to it's hype.

I acknowledge that AMD chips are cooler and faster and a better value, and recommend them to people all the time, but I still run Intel chips all the time myself.

I am not waiting for Conroe at all. Certainly not dying for it.

When it comes, it will just be the latest Intel chip to me.

I never can figure out all the vitriol over who runs what.

Some folks like Hondas, some like Fords, some like Dodge, etc.
Some folks think they're all just cars....

I had plans to build AMD systems, but circumstances intervened and I just never got around to it.

AMD is better, and I run Intel. Get over it....

yeah.. thats my opinion as well..
I run Intel.. big whipdee dooo... It runs a little hotter... another big whipdee doo..
Ok.. so I'll pay another $1 or $2 in electric costs over a years time.. easy there.. better start saving some pennies

Some forum members try to get too much into the technical details of it all.. when everyone should really just enjoy the technology that we have