Topic Title: Difference between 68W and 95W TDP Gen1 Dual-Core Opteron 285?
Topic Summary: The Dual-Core Opteron 285 comes in 68W and 95W flavors w/same performance. How is this possible?
Created On: 01/16/2013 06:33 PM
Status: Post and Reply
Linear : Threading : Single : Branch
Search Topic Search Topic
Topic Tools Topic Tools
View similar topics View similar topics
View topic in raw text format. Print this topic.
 01/16/2013 06:33 PM
User is offline View Users Profile Print this message

Author Icon
sudos
Peon

Posts: 3
Joined: 01/16/2013

 

I'm currently using two of the retail-grade 95W TDP Opteron 285's in this set-up. My current dilemma is temperatures. I have two Zalman CNP5X heatsinks on these CPUs, but they barely do the job, as you can see, to keep these yesteryear CPUs cool, or at least, as cool as I'd like them to be (high 20C to very low 30C range.)

Speccy Temps.

While looking around, I came across model number OST285FAA6CB on CPU-World for a 68W TDP 285, and further research shows that it's an OEM part for an old generation of the HP DL385 line of servers. Also according to CPU-World, and other sources that I can find, I can't find any difference between the 95W and 68W versions besides the TDP... which I find very peculiar seeing as they're both 90nm CPUs that seem to otherwise be identical besides power consumption. I can get a pair of the 68W 285's for relatively cheap on some websites, and this would be ideal if this actually means I'll be using less power and my temperatures will go down to the range I'd like them to be while idle.

My question is, what's the difference that makes the 68W Opteron use less power than the 95W Opteron, if they are both of the same 90nm SOI process type and both perform the same on paper? is it a matter of stepping? were they hand-picked and marked based on power usage for deployment in HP servers? Were miniscule improvements made to the cores that decreased power consumption and just weren't documented? and although they're the same, as said, on paper, will I theoretically see any slight improvement in performance?

Thank you very much in advance.

 

 01/16/2013 08:54 PM
User is offline View Users Profile Print this message

Author Icon
black_zion
80 Column Mind

Posts: 12209
Joined: 04/17/2008

I see two Opteron 285 models, a 95w (OSA) and an 85w (OST) version. I'm not sure you will see any difference due to the way AMD rates their chip's TDP, which is the maximum it could ever achieve. My advice would be to upgrade your cooling, or upgrade your system to a single modern Opteron setup.

Or live with the higher temps, for a 90nm CPU 40s isn't high.

-------------------------
ASUS Sabertooth 990FX/Gen3 R2, FX-8350 w/ Corsair H60, 8 GiB G.SKILL RipjawsX DDR3-2133, XFX HD 7970 Ghz, 512GB Vertex 4, 256GB Vector, 240GB Agility 3, Creative X-Fi Titanium w/ Creative Gigaworks S750, SeaSonic X750, HP ZR2440w, Win 7 Ultimate x64
 01/16/2013 10:06 PM
User is offline View Users Profile Print this message

Author Icon
Mime
Forum Moderator

Posts: 578
Joined: 08/28/2012

OST is the prefix for Opterons sold as OEM for HP.  It's got the same Revision code and the same CPUID as the OSA285, so I'm thinkin the reason why it got a different rating for TDP is probably less technical and more political.... Meaning, HP told AMD they wanted a TDP of 68W and AMD "found" a way to make it happen.

TDP is somewhat arbitrarily defined anyway.  It isn't a rating of how hot or how much power your chip will consume, but only defines the envelope where it will operate.  Even with a different TDP, if there's no actual differences in process, voltage or frequency, then there's not likely to be much change in temperature.

There are HE Opterons which are chips that were cherry picked to operate with a lower vcore, meaning lower power consumption and cooler temperatures, but they're often quite a bit more expensive.



-------------------------

Do not meddle in the affairs of archers, for they are subtle and quick to anger.
Post Count: +8510
Troll Hunter

The opinions expressed above do not represent those of Advanced Micro Devices or any of their affiliates.



Edited: 01/16/2013 at 10:12 PM by Mime
 01/17/2013 01:33 AM
User is offline View Users Profile Print this message

Author Icon
sudos
Peon

Posts: 3
Joined: 01/16/2013

Originally posted by: black_zion I see two Opteron 285 models, a 95w (OSA) and an 85w (OST) version. I'm not sure you will see any difference due to the way AMD rates their chip's TDP, which is the maximum it could ever achieve. My advice would be to upgrade your cooling, or upgrade your system to a single modern Opteron setup. Or live with the higher temps, for a 90nm CPU 40s isn't high.

Yes, but the temperature is still high up enough at load (55C, almost 60C) under stress that it's just a nuisiance. actual hardware upgrades are out of the question, I spent too much time building this already. As far as a cooling upgrade, I'm still researching alternatives. the CNP5X heatsinks I'm currently using are exactly what I need in terms of the direction of airflow and the mounting capability. I'm having a problem finding a heatsink now that has two latches that can sit atop the standard AMD bracket, or use the two screw holes meant to mount such bracket to the backplate.

To give you an idea of my heatsink size constraints in this case, here's an image: [http://goput.it/msm.jpg]

(and as a note for the record, I'm working on acquiring heatsinks for the MOSFETs for both CPUs. all of the S939 boards I've come into contact have sinks on them, so I don't know why an S940 board should be any different... probably since it was half-made to go into a high-airflow server/workstation-class case.)

 If a heatsink can be found on the cheap that isn't similar at all to the stock OEM heatsink, and can handle a 95W+ load, I'd switch to them in a heartbeat.

 

Originally posted by: Mime OST is the prefix for Opterons sold as OEM for HP.  It's got the same Revision code and the same CPUID as the OSA285, so I'm thinkin the reason why it got a different rating for TDP is probably less technical and more political.... Meaning, HP told AMD they wanted a TDP of 68W and AMD "found" a way to make it happen.

I was figuring it would be something along those lines, it just seemed too good to be true.  So, bottom line on that is, based on that educated guess, there is no difference at all between the OSA and OST Opteron 285 regardless of the TDP rating given to it... would it be correct to infer at least that much?

 01/17/2013 01:13 PM
User is offline View Users Profile Print this message

Author Icon
Mime
Forum Moderator

Posts: 578
Joined: 08/28/2012

Yeah I'd say that's probably a safe conclusion.  It's hard to tell exactly what the difference is, if any.  Maybe they're just binned differently, sort of like the HE chips, but without the same drop in voltage.



-------------------------

Do not meddle in the affairs of archers, for they are subtle and quick to anger.
Post Count: +8510
Troll Hunter

The opinions expressed above do not represent those of Advanced Micro Devices or any of their affiliates.

Statistics
84179 users are registered to the AMD Support and Game forum.
There are currently 3 users logged in.

FuseTalk Hosting Executive Plan v3.2 - © 1999-2014 FuseTalk Inc. All rights reserved.