Overclocking Extravaganza: Radeon HD 4890 To The Max
by Derek Wilson on April 29, 2009 12:01 AM EST- Posted in
- GPUs
Combined Memory and Core Overclocking: The Sweet Spot
In this round of tests, we combine our previous maximum overclocks. This is our compromise, in that we show the maximum potential of combined core and memory overclocking rather than effects of memory overclocking over each core clock speed we tested. While the latter option would be more complete, our tests do enough to show people what they need to know to find the sweet spot.
We theorized that with an extreme core clock speed that memory may have become a bottleneck to performance at some point. Despite the fact that increasing memory clock without increasing core clock didn't do much at all, we could see increased benefit beyond what one might expect based on our initial memory overclocking results.
Before we looked at varied memory clock with a stock core clock and varied core clock with a stock memory clock. Let's revisit both of those but also add in a twist. We will also look at percent increase in performance when overclocking memory with a 1GHz core clock and the percent increase in performance when overclocking the core with a 1.2GHz memory clock.
1680x1050 1920x1200 2560x1600
1680x1050 1920x1200 2560x1600
1680x1050 1920x1200 2560x1600
1680x1050 1920x1200 2560x1600
Note that in both cases we see a much bigger boost in performance. This means that while applications tend to be very heavily compute limited, at higher core clock speeds on AMD hardware memory bandwidth increasinly becomes a bottleneck. Now let's take a look at what we get when going from a completely stock part to a maximally overclocked part at 1GHz/1.2GHz (core/mem).
1680x1050 1920x1200 2560x1600
To get a basic idea of what's going on, here's an example of two programs. Remember that this isn't really real world and is just to illustrate the concept.
The first application is completely compute bound and the second is 50% compute bound and 50% memory bandwidth bound. Both tests generate 100 frames per second on a stock Radeon HD 4890. If we increase core clock speed 10%, the first application will generate 110 frames per second, while the second one would only generate 105. This is because we only see the 10% benefit while doing half of the work. If we look at only boosting memory performance 10%, the first program delivers only 100 fps while the second hits 105 again. Pushing both memory and core clock speed up 10% each gives us 110 frames per second from both applications. Basically.
Nothing is really that contrived or works like that, but the important thing to remember is that different applications can make varying use of different resources, and balancing those resources is important to ensuring the best performance in the most efficient package.
So, to find the sweet spot for your overclock, you will want to increase core clock speed as much as you can. Then bump up memory clock and see how high you can get it and remain stable. Use a real world application to test performance at each point and then use a binary search like algorithm to find the sweet spot in a short number of tests. And there you have it. We didn't do this for you, but what's better practice than a little hands on experience right? Besides, if gives readers the opportunity to compare notes in the comments on what the optimal memory clock for a 1GHz core clock on the 4890 would be. Have fun!
Facebook
Twitter
RSS
61 Comments
View All Comments
PrinceGaz - Wednesday, April 29, 2009 - link
Interesting certainly, but when I read Overclocking... to the max" my first thought that a large tub of LN2 would be used :pDerekWilson - Wednesday, April 29, 2009 - link
heh ... i'd love to play with some liquid nitrogen. that'd be great fun. i don't think we're going there anytime soon though.Veteran - Wednesday, April 29, 2009 - link
Why didn't you guys play with the fanprofiles? Added some VGPU with 3th party software? Just went to the realy maximum of the card. It would be amazing to see the results of that. Using 3th party software shouldn't be any limitation to overclocking if you ask me.DerekWilson - Wednesday, April 29, 2009 - link
We did set the fan on 100% manually all the time. Sure, it's a leafblower at that speed, but we wanted to keep it cool. Sorry I didn't mention that.It is possible to get some 4890 cards higher, but this wasn't an article for the hardcore overclocker. Rather we wanted to talk about what the average user can do out of the box by clicking a few buttons in the driver. Which, in my opinion, is more impressive than if we'd gotten an extra couple percent increase in core and memory clock speeds.
Not that even I wouldn't be interested in knowing how fast you could get one of these beasts ... It's just a different article than the one I wanted to write here.
SiliconDoc - Saturday, June 6, 2009 - link
It will be a nice day when you don't have to apologize ten times in every ati vs nvidia article for your massive ati biased BS Derek.--
" We did set the fan on 100% manually all the time. Sure, it's a leafblower at that speed, but we wanted to keep it cool. Sorry I didn't mention that. "
---
All you red roosters aren't sorry - the enthusiasts see your charts and head off to redland purchse while your lying CRAP misdirects them ! And you know it - and are sickeringly proud of it no doubt !
It's been going on forver by you here !
PERIOD!
Veteran - Wednesday, April 29, 2009 - link
The title just imply different things then this review. That's all, i enjoyed reading it and see how the card responds to Core and Mem OC's. But if you use the words 'extravaganza' (very special) and 'to the max' people start thinking about dry-ice/ln2, voltage modifications etc.The article is awesome, the title is a little bit wrong
CyNics - Wednesday, April 29, 2009 - link
to Derek: have you tried overclocking the GeForce GTX 275, or at least comparing an OC'ed gtx275 with an OC'ed 4890 before coming to such conclusion?DerekWilson - Wednesday, April 29, 2009 - link
This is a pretty good question --These tests and the article took a while for me to put together. I am planning on doing a GTX 275 overclocking and seeing if anything interesting happens. If I get good results I'll put together an article on the subject.
I tried pretty hard to keep this article on the subject of the 4890 itself and where I made comparisons it was to hardware outside it's price class with the GTX 285.
No matter how well the GTX 275 does, the 4890 still did as well as it did versus the GTX 285 (and relative to itself).
If I screwed up and made direct comparisons between overclocked 4890 and GTX 275 or something, let me know and I'll fix it 'cause I didn't mean to. Otherwise I hope that this article is as appropriate as I wanted it to be.
Once I've also overclocked the GTX 275 I'll have something to say about it's relative value versus the 4890 in that regard.
SiliconDoc - Saturday, June 6, 2009 - link
THAT'S AN AMAZING BIT OF BS DEREK !" If I screwed up and made direct comparisons between overclocked 4890 and GTX 275 or something, let me know and I'll fix it 'cause I didn't mean to. Otherwise I hope that this article is as appropriate as I wanted it to be.
Once I've also overclocked the GTX 275 I'll have something to say about it's relative value versus the 4890 in that regard. "
First you say IF you used a non overclocked GTX275 and screwed up doing it - someone should point it out (because you obviously can't see it in your own charts - right red rooster?) - and if someone points out your massive idiots bias, you'll change it otherwise you won't... THEN YOU ADMIT YOU HAVEN'T OVERCLOCKED THE GTX275 YET ! PROVING ABSOLUTELY YOU DID IN FACT COMPARE IT NON OVERCLOCKED TO THE 4890 SPECIAL MANUFACTURERS EDITION SENT TO YOU !
-
SO WHAT THE HELL DID YOU MEAN BY " If I screwed up and made direct comparisons between overclocked 4890 and GTX 275 or something, let me know and I'll fix it 'cause I didn't mean to." ?!!???
--
YOU DID IT DEREK - YOU KNEW YOU DID IT, YOU DID IT "ON PURPOSE" THEN EVEN AS YOU DENY IT AFTER BEING QUESTIONED, YOU ADMIT IT INCRIMINATING YOURSELF !
YOU EVEN ASK THE QUESTONER POINT IT OUT WHEN YOU DID THE CHARTS AND THE TESTS HERE !
MY GAWD!
--
Dude, if you think your BS is even passable, THINK AGAIN !
SiliconDoc - Saturday, June 6, 2009 - link
Yes, our great masterous "fair" and "unbiased" reviewer Derek.
DOESN'T EVEN KNOW WHEN HE IS COMPARING OC'ED SPECIAL MANU. TO STOCK !
.
.
WHAT a genius! Keep that boy hired ! Must be worth a whole 50 cents an hour, if that !