Budget Battle: HyperMemory vs. TurboCache

33 Comments

View All Comments

• OrSin - Thursday, May 12, 2005 - link

  Why test $60 video cards on systems with the highend chips and memory. No one test goodyear tires on ferrari. I want to see these test ran on 2800 CPU and kinmax memory. As it stands this a waste to me.
  
  Also why not test some on board cards to these.
  To see if its even worth upgrade the Intel ,ATI, or NV solution to these.
  
  

• stevty2889 - Thursday, May 12, 2005 - link

  What the heck? The system it was tested on was:
  Microsoft Windows XP SP2
  ASUS A8N-SLI Deluxe
  AMD Athlon FX-53
  1GB OCZ PC3200 @ 2:2:2:9
  Seagate 7200.7 HD
  OCZ Powerstream 600W PS
  
  Nobody that buy's these cards is going to be running on a system like that..should have been tested with a sempron, or 2.8ghz P4/A64 2800+ type setup instead..

• DerekWilson - Thursday, May 12, 2005 - link

  Sorry about leaving off the machine specs -- I've updated the article.
  
  Actually, that was quite an oversight as the system these cards are run in is very important to note when looking at the numbers.
  
  Marlin1975, InuYasha is correct -- the 32MB card out performed the 64MB part in almost every test. It wasn't until we upped the resolution to unplayable degrees that the 64MB part was able to make up the difference.

• bob661 - Thursday, May 12, 2005 - link

  Does anyone know what motherboard and how much ram was used in their test system?

• CrystalBay - Thursday, May 12, 2005 - link

  #3 I was wondering the same thing myself, hmmm.

• InuYasha - Thursday, May 12, 2005 - link

  #4 the 32MB and 64MB results are not backward.
  
  if i remember correctly from HOCP, the 32MB has faster memory and that makes a huge difference than the amount of memory.

• Icehawk - Thursday, May 12, 2005 - link

  I don't see the machine specs anywhere either? I'm curious if these were tested on the "standard" uber-machine or tested on what kind of PC someone buying these would actually have. Somehow, by the #s generated I think this was on the uber-machine. While interesting to see ultimate performance I think end-users would also be served by showing more realistic performance #s.
  
  A couple of minor typos ;)

• Marlin1975 - Thursday, May 12, 2005 - link

  OK I think it is a mis-tpe, not the graohs that are wrong on page 7...
  
  "Unreal Tournament 2004 shows our 32MB HyperMemory performing on par with the 64MB TurboCache part in the middle of the pack. The 16 and 32 MB TC cards round out the bottom and top of the pack respectively."

• gimpsoft - Thursday, May 12, 2005 - link

  be alot better if it could be the other way around taking video card memory and using it for windows then ill really pay for it lol letting video card borrow system memory bad idea for the future
  
  anyways that's just me

• Marlin1975 - Thursday, May 12, 2005 - link

  Do you have the 32mb and 64mb cards backwards?