• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

AM2 4800+ benchmarked by THG

Is the RAM really running at DDR2 667? PC Mark's and Sandra's memory benchmarks don't see an increased bandwidth... which is the only good thing about DDR2 performance-wise.
 
Originally posted by: Furen
Is the RAM really running at DDR2 667? PC Mark's and Sandra's memory benchmarks don't see an increased bandwidth... which is the only good thing about DDR2 performance-wise.
Click to expand...

I was wondering that too.
 
With a updated and more mature BIOS, plus a enthusiast motherboard like DFI, then the AM2 should be slightly faster.

Kinda disapointing, hopefully THG missed out on some other benchmarks that showed AM2's full potential. But i think ill stick to my socket 939, until end of year when 65nm AM2 processers come out.
 
I suppose the performance was a bit dissapointing but then THG did say that its not a final production cpu that was tested.

DDR2 will be helpful in the future. Those future quad-core cpus will definiately need it. In fact I read an article a few months back on either anandtech or xbitlabs (can't remember), but anyway it showed that in some applications like video encoding, DDR400 was already starting to bottlneck the X2 cpus.

The power reductions though do look nice. 🙂
 
Looking at the Sandra memory benchmarks it is obvious that the full memory bandwidth isn't being used. Sandra's test isn't sensitive to timings either, so it isn't just the bad timings of the DDR2 that are at play here. I'd say that the memory isn't really running at 667MHz at all, it's probably at 400MHz. This explains why the AM2 system performs worse than the S939 systems in most tests.

Another possibility is that there is in fact a bug in the memory controller of these early chips. Either way, it's quite clear that this isn't the kind of numbers we'll see at the official launch. That said, I wouldn't explain any miracles either... 🙂
 
Good to see the platofrom is stable and running on current chipsets. It's also good to see that AMD can afford to wait for the right technology to be widely available before switching. The fact that Intel need to drop to 65nm to get 65W TDP on Desktop parts while AMD can do it on 90nm is a signof the times.

Good work to the guys at AMD and lets hope DDR2-800 will push it ahead of DDR-400 by a little bit at least.

Edit: The RAM was running at slower timings aswell, instead of CAS3 it was CAS4 but that won't affect the scores too much.
 
Originally posted by: cmrmrc
http://www.tomshardware.com/2006/02/21/a_look_at_amds_socket_am2_platform/

It seems that DDR2-667 isn't helping AMD for now..as the AM2 4800+ is outperformed in almost every test by the 939 4800+...

let's wait and see if DDR2-800 will help a bit...
Click to expand...

Ya but it also is runnong on horrible timings. Lets wait for the retail release to figure things out 🙂 Remember, anand posted that his chip and mobo had problems, and he was waiting for a better engineering sample before he would post results?
 
Don't forget that the reason that the socket AM2 is delayed is that there was a DDR2 controller bug that reduced performance that is being fixed. This pre-production CPU probably has the DDR2 bug.
 
Originally posted by: Nyati13
Don't forget that the reason that the socket AM2 is delayed is that there was a DDR2 controller bug that reduced performance that is being fixed. This pre-production CPU probably has the DDR2 bug.
Click to expand...

ya thats what I meant 🙂
 
I wouldn't say that performance "blows" though, so perhaps this is a motherboard issue more than an issue with the CPUs.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top