The Radeon HD 4870 1GB: The Card to Get

78 Comments

View All Comments

• poohbear - Sunday, September 28, 2008 - link

  "The best example of this is the 8800GTS which I assume many of us still own. We don't care how the 4870 runs on the Intel Core 2 Extreme QX9770 @ 3.20GHz used in the review, we care if the card will boost our FPS enough to warrant purchasing for our computer. It's a different type of comparison than the GPU-limited and CPU-limited tests they currently run, but very useful to the majority of us."
  
  QFT
  
  please anandtech show us graphs that tell us if its worth it to upgrade from an 8800gt or similar hardware, that's what the majority of us own and are interested in knowing. This 1gb 4870 is interesting and all, but why would you do an article like this before addressing the aforementioned question that is pertinent to 90% of the rest of us?
  

• PrinceGaz - Friday, September 26, 2008 - link

  I must concur with this guy.
  
  Whilst I am more than capable of internally extrapolating other test results to these latest ones, I have an 8800GTS (640MB) so with all reviews on AT have to go through a double compare of the closest in the review to what I have.
  
  Please Anand & co, include at least high-end cards from nVidia and AMD from two generations ago in all graphics-card reviews, and mid-range cards from ther previous-generation where relevant. We don't all buy a new card every few months, most people will buy a new graphics card every two years or less often, even quite a few of us who visit this site every day.
  
  When I only see comparisons to other current cards as in this review, I invariably abort reading it and find a review on another site which includes a wider range of cards, which is a shame as I enjoy the technical info here- but if the only comparison provided is between current generation competitors, neither of which I have, then all the graphs are pointless as I (and most other people) have previous generation cards.

• SiliconDoc - Friday, October 3, 2008 - link

  That would be imformative and then we'd be informed. In tihs case, you must be a $$$$$$$$$$$$$$$$$$$4 loaded supergame monster - and so it doesn't matter.
  Just crank up another 4 or 5 hundred on the collegiate mastercard if you blow the first couple of choices...you know that can be the geekfest lan machine or whatever.

• saiku - Wednesday, October 1, 2008 - link

  I must also agree to the two posters above me. I too have a 8800 GTS and would like to know if this card makes sense for an upgrade. Can AT please keep 1-2 year old tech around when they talk about new parts?

• formulav8 - Saturday, September 27, 2008 - link

  Anandtech finally game AMD the props they deserved along time ago instead of kissing up to nvidia who had no care in the world about charging almost a literal arm and leg. Not that AMD/ATI cares all that much but at least they didn't milk us for all we're worth like nVidia did the past couple years. Just my opinion of course :)
  
  Decent review by the way. I do agree though that reviews should be more real world in the sense that they be reviewed with more reasonable, mainstream components like a Phenom, Higher End X2, and Low-Mid end Core2, memory configs like 2GB at 667/800 mhz, ect... Since that is where most gamers will be at. Again, just my opinion :)
  
  
  Jason

• SiliconDoc - Saturday, December 27, 2008 - link

  How about mainstream monitor resolutions ?
  Oh that's right , the highend cards blow away mainstream monitor resolutions now.
  Well, I guess they had better start including the 2 grand required for proper monitor upgrades to run the resolutions they post all their reviews in now.
  Somehow that giant 30 inch Dell DWP (or two or three) 2560x1600 sitting in the test lab isn't magically appearing attached to their readers gaming rigs.

• Jorgisven - Monday, September 29, 2008 - link

  The trouble with using mid-range components is that they can unneccessarily bottleneck the performance in ways that can affect different GPU's differently, in different situations. If you scale back your components and then put a $500 GPU in, the results would be skewed, especially at higher details. Since this is a higher end GPU, one would expect that you would have components relative in price as well.

• SiliconDoc - Friday, October 3, 2008 - link

  That's not the worst of it Jordan ...
  How often do you run a 2550x 1900 or whatever supermegga massive giganto resolution ?
  I guess I'll have to take a rtip to Taiwan for my 37" super rezz megoo mighty mouse monitor... that's only $2,567.98 plus tax and overseas shipping to even run the ding dang rezzzzzzzzzzzzzes they test in nowadyas.
  Yeah, gee, we've all got these monitors that run this extraordinary resolution that you CANNOT EVEN FIND in the retail store market - and barely online when you try.
  CRIPES.
  ( A friend of mine can't stop cranking up his rezzz - he just got 2 more monitors last night - so he cranks the thing up till the monitors are buzzing and hissing LOL - then I send him a pic or a screenshot or something - and say blah blah - and he can't read the thing - the lettering is the size of the ding dang microdot.
  rofl
  .000000035 um lettering. lol
  He doesn't READ much. rofl.

• Patrick Wolf - Thursday, September 25, 2008 - link

  While you can go back to previous articles which have basically the same Test Setup with the same games, it would be very convenient if they would include cards from the 9 series in their newer articles.

• Hxx - Thursday, September 25, 2008 - link

  Why should they be compared? 9800s are old tech, not in high demand anymore.Thats why
  
  About the article, the writer did a very good job. I'm surprised to see the 4870 coming up in front of the gtx280 in some games. The 4870 1gb is the best card from a price/performance perspective.
  Looks like Nvidia is due for another price drop, lol. Good job ATI.