AMD 90nm power consumption measured

EugeneMc

Distinguished
Dec 22, 2003
55
0
18,630
At Tech Report:

<A HREF="http://techreport.com/onearticle.x/7417" target="_new">http://techreport.com/onearticle.x/7417</A>

It's a successfully transition. In the other hand, Prescott seems more horrible than ever.
 

RichPLS

Champion
So true...


========================
Try everything...
Do not be afraid of failure, for this is how we learn and grow...
Live life to the fullest...
Do not regret what you have not yet done!!!
 

trooper11

Distinguished
Feb 4, 2004
758
0
18,980
im relieved to see that all the fear from taht first test everyone saw was not the real story it seems. i wait to see more numbers, but thsi looks much better.
 

phial

Splendid
Oct 29, 2002
6,757
0
25,780
mmm that makes me horny

-------
<A HREF="http://www.albinoblacksheep.com/flash/you.html" target="_new">please dont click here! </A>
Brand name whores are stupid!
 

Crashman

Polypheme
Former Staff
I knew a Scotty who always had a Woody for a Willy.

<font color=blue>Only a place as big as the internet could be home to a hero as big as Crashman!</font color=blue>
<font color=red>Only a place as big as the internet could be home to an ego as large as Crashman's!</font color=red>
 

darko21

Distinguished
Sep 15, 2003
1,098
0
19,280
I'm still waiting for the 5% performance gain. I wonder when we will see that. I agree the thermal properties look ok to good for now.

If I glanced at a spilt box of tooth picks on the floor, could I tell you how many are in the pile. Not a chance, But then again I don't have to buy my underware at Kmart.
 

Cybercraig

Distinguished
Dec 31, 2007
2,058
0
19,780
I'm still waiting for that guy to be able to afford air-conditioning. Who the hell can get anything done in an office at 85F?

Abit IS7 - 3.0C @ 3.6ghz - Mushkin PC4000 (2 X 512) - Sapphire 9800Pro - TT 420 watt Pure Power
Samsung 120gb ATA-100 - Maxtor 40gb ATA - 100
Sony DRU-510A - Yellowtail Merlot
 

El_Jefe_77

Distinguished
Aug 10, 2004
141
0
18,680
Yeah, that is a mystery right there.
Also is a mystery is the FACT that this guy is the ONLY ONE ever to say what he has said about thermal things.

Anyone else note the:
Update: Cool'n'Quiet was not enabled on the Socket 939 motherboard.

What does that mean? How does that effect the tests? Can a 3800 clocked down to 3500 on a 130 be equal to a 3500 in every way for a test comparison? Does he have a standardized method for measuring cpu temps? What thermal grease did he use? how often has he done tests? Any other battery of examples or tests for us to see? How does cool and quiet work in an idle state?

My understanding is this: Cool and quiet wasnt enabled on 939, but no mention of it on 754. I know the ABIT board im about to purchase, kv8 pro, has cool and quiet enabled, and certainly would have less wattage usage in the idle state than shown in his test. An assumption, perhaps, but a fairly decent one. THis guy was nice in showing his test, but he failed to make it a real study of the thing.
 

MaximumGoat

Distinguished
Jul 22, 2003
115
0
18,680
"Cool'n'Quiet was not enabled on the Socket 939 motherboard.
What does that mean?"
if CnQ was enabled then the athlon64s would have had dramatically reduced power consumption in idle (as they clock down to 1ghz and lower voltage), however it would make no difference in the other tests.

"Can a 3800 clocked down to 3500 on a 130 be equal to a 3500 in every way for a test comparison?"
this is debatable, but essentially yes, and even if it isnt fair then it only makes the new 90nm process better as a clocked down 3800 is probably more efficient than an original 3500.

"Does he have a standardized method for measuring cpu temps?"
this makes no difference as long as he uses the same method for both systems, and as it was the same system for both tests but with the cpu swapped i think this is a fair assumption. he doesnt mention the temp of the P4 so there is no comparison problem there

"What thermal grease did he use?"
again, as long as it is the same in all tests then makes no difference, and as he measures power not temperature (well ok he does measure temp, but only in passing) then it really doesnt matter.

"Cool and quiet wasnt enabled on 939, but no mention of it on 754"
there is a reason he doesnt mention the 754, they are both 939!


"THis guy was nice in showing his test, but he failed to make it a real study of the thing."

actually it is a really good article, not perfect, but very good. you failed to study properly.

.
 

El_Jefe_77

Distinguished
Aug 10, 2004
141
0
18,680
I have seen hundreds of articles on cpu temps. Many of them, those with more than a half of page of information, discuss the inability to standardize their temperature readings because the sensors are not placed in the same locations, and upon finding a better way to measure temperature, find different results than at first. The variability in cpu temps is horrendous between articles, and while differences may be the key to understanding, temperature trully is itself a standardized measurement, and differing readings from differing websites leans more towards human error than some new found glimmer of hope that one site is reporting.

If these things are cooler than 130 nm, why has every review besides this one said otherwise??!!
 

MaximumGoat

Distinguished
Jul 22, 2003
115
0
18,680
the review primarily focuses on power consumption not temperature, the author even says not to take the results as gospel.

it is suggested that even if the 90nm core uses less power it could still get hotter as it is a much smaller core to disipate the heat. but of course this doesnt explain how he found lower temps when others havent.

my answer to this is either:

1) cores (especially new ones) are not all equal other sites may have had older cores and therefore got different results.

2) other sites "data" is based on speculation that 90nm is inherently more power hungry than 130nm, which is in turn based on intels 90nm cores.

this is the most reliable artical i have found,

also note power is the power of a whole system (- monitor) not just the cpu, it the posts below the article many seem supprised a cpu can draw over 200watts!!

.
 

ChipDeath

Splendid
May 16, 2002
4,307
0
22,790
erm. It's the same motherboard, in the same office, using the same software package to read the same on-board thermistor, using the same Heatsink applied by the same person.

It doesn't matter how innacurate his measuring is, it's going to be equally innacurate for both setups, so a recorded drop in temperature is a valid one, in this case.

A Question though: Does the 3800+ run at a higher voltage than the 3500+ (by default)? If that's the case and he didn't reduce the voltage along with the clock speed then that would account for the temperature disparity not seen elsewhere. Just a thought.

---
Epox 8RDA+ V1.1 w/ Custom NB HS
XP1700+ @200x10 (~2Ghz), 1.4 Vcore
2x256Mb Corsair PC3200LL/1x512Mb Corsair XMS PC4000 2.5-3-3-7
Sapphire 9800Pro (VGA Silencer Rev3) @418/742
 

Schmide

Distinguished
Aug 2, 2001
1,442
0
19,280
I just enabled cool and quiet on my 3500 939 90nm and it’s amazing. It under clocks the processor to 1ghz at 1.1v (5x200) and along with the smart fan cpu target function, it stops the CPU fan and it idles at 40-45c. Normally I’d freak out but it seems to have its crap together under this function and starts the fan up nicely if it gets to hot. I can surf the web and even do minor graphics (watch small videos) and it remains cool and quiet.

(Motherboard MSI K8N Neo2 Platinum)

Dichromatic for your viewing plesure...
 

Worf101

Distinguished
Jun 25, 2004
498
0
18,780
Works for me. When combined with my Thermaltake Silentboost K8, the Raptors make more noise than the fans do. It's a beautiful thing.

Da Worfster

If you don't stand for something, you'll fall for anything.
 

trooper11

Distinguished
Feb 4, 2004
758
0
18,980
also, correct me if im wrong, but i didnt see him say he was using anything other then the packaged heatsink/fan for the athlon so he wouldnt be using any thermal grease, just the packaged thermal pad, but if he is changing them out then he did have to use grease, but it doesnt really matter as long as both get the same treatment.
 

tombance

Distinguished
Jun 16, 2002
1,412
0
19,280
I seem to be the only one who cant get CnQ Working. Im using an MSI K8N Neo Platinum with a 2800+ and the retail HS and Windows 2000. I've enabled CnQ in the BIOS (latest Version) and tried installing both the driver from the CD and the latest one from AMD's website but it still doesnt work. Any suggestions?