CrossFire Xpress 3200: RD580 for AM2

71 Comments

View All Comments

• Wesley Fink - Thursday, June 1, 2006 - link

  1X Increments corrected.
  
  We did not have audio performance data for nVidia chipsets in the 590 launch review, but it will be included in our roundup of 6 AM2 boards which is in process. I have added numbers for the Foxconn ( nForce 590) HD codec for reference. Foxconn is the nVidia Reference board.
  
  The board photo was captured at 12 Megapixels. Unfortunately, the "Save for Web" feature in Photoshop which gets the image to a reasonable file size for posting a 1280 image compromises sharpness at higher resolutions.

• Trisped - Thursday, June 1, 2006 - link

  Thanks for the Foxconn numbers.
  
  So you used "Save for Web" and lowered the quality so it would be easier to download? That makes sense. A 43k file is much better then a 1M one.

• JarredWalton - Thursday, June 1, 2006 - link

  Or 422K vs. 5+ MB. ;)

• lopri - Thursday, June 1, 2006 - link

  I truly appreciate AT staff's responses to my questions. It cleared so many things that I questioned while reading the review, so now I'm understanding better.
  
  

  quote:
  

  ---

  the fact is DDR400 was the fastest memory standard for DDR. Anything higher was overclocking. For DDR2, we have DDR2-800 as the current highest standard speed

  ---

  
  
  This is actually the only possible explanation that I could think of. You're right in that DDR400 is the fastest JEDEC approved speed. I sort of guessed but still, considering the ammount of memory reviews you've done in the past, thought a bit stranage. But thank you for explaining. Request, however: Please do a out-of-the spec DDR vs DDR2 reviews in the future. :D This can be a big factor for people who actually consider upgrading.
  
  

  quote:
  

  ---

  AMD introduced this platform with very conservative timings and tables for the board and memory suppliers to follow. We expect to see 1T timings at 800 later this year as AMD "massages" the memory controller. I ran tests at DDR2667 1T and they were basically the same or slightly worse than DDR2800 at 2T with all other settings being equal. The problem is we cannot run tRP and tRCD lower than 3 currently so any advantage of 1T is being wasted due to higher latencies. On a couple of our review boards we could also run DDR2800 at 4-4-4-15 1T but the 3-3-3-13/9 2T setting provided better memory bandwidth and lower latencies overall. We are still testing various memory settings as each board has been a little a different in optimizations made by each supplier. We will have a separate review on EPP and Memory settings for AM2 in the near future.

  ---

  
  
  Again, I appreciate the explanation. Not knowing about DDR2 much myself still, I could not have known it when reading the review. It'd have cleared up some misunderstanding if you have mentioned the 1T/2T issues in the review (like above), it'd have helped a ton to understand. I'm sure there are many different traits of DDR2 compared to DDR, without such knowledge I could not help but questioning. Thank you, Gary.
  
  Still the 1T/2T issue on AM2 is somewhat disappointing. (Not reviewers' fault) I have a bad feeling that AMD's IMC won't be able to handle 1T for DIMMs faster than DDR2-800, even with future revision. :( For entire lifespan of Socket 939, they couldn't get 4 sticks to run @1T timing.. (except a couple going-around of DFI's)
  
  
  
  

• lopri - Thursday, June 1, 2006 - link

  quote:
  

  ---

  Our Corsair or OCZ PC8500 sticks will run at 3-3-3-9 2T at 800 with a small voltage increase to 2.2V easily although the memory is rated at 5-5-5-15 2T. I am working on a single versus dual channel DDR2 article at this time, cutting to the chase, single channel DDR2 with fast timings will provide up to 98% of the performance of dual channel DDR2 under the same conditions. It might be something to think about when looking at $350~$500 DDR2 2GB kits.

  ---

  
  
  Also if this is true, it's an absolutely fantastic news. Please let us know the detail as soon as you can. Thank you.
  

• DigitalFreak - Thursday, June 1, 2006 - link

  quote:
  

  ---

  ATI did not need to develop a new chipset for the new Socket AM2. Why then has it been so difficult for ATI to have AM2 chipsets ready for launch?

  ---

  
  
  Maybe they didn't need to develop a new North bridge, but the South bridge is another matter. With ULi supplies drying up, it would have been extremely stupid to use the SB450 yet again.

• Myrandex - Thursday, June 1, 2006 - link

  It was stated that the ATI solution was better tahn the ULI and less than Nvidia, however in the graphs it was less than both, although very close to ULI.
  Jason

• Wesley Fink - Thursday, June 1, 2006 - link

  The statement is correct. Going back to review notes there was a typo in the chart creation which has now been corrected. USB throughput for SB600 is 241.6 and not 231.6 as shown in the earlier chart.

• Alyster - Thursday, June 1, 2006 - link

  I just wonder if SB600 will be available on 939 boards in future. I'm going to purchase ATI based MSI-RS482M4-ILD mATX motherboard with SB450 and may be I should wait untill they start offering SB600 on mATX boards. Any suggestions? Thanks

• Wesley Fink - Thursday, June 1, 2006 - link

  As we understand it, SB600 is not pin-compatible with SB450, so it is not a drop-in for the older chip. We therefore think it is unlikely you should wait for a board redesign on an older 939 board. Any new 939 boards - and there may be some if the market wants them - will likely use SB600.