Discussion in 'Reviews' started by W1zzard, Apr 9, 2010.
To read this review go to: http://www.techpowerup.com/reviews/HIS/HD_5850_iCooler_V_Turbo/
Excellent review as always W1zzard, keep up the good work. I would like to see a review featuring the GTX 470 in SLI. Pretty please? XD
nice review w1zzard, i was wonder on the HAWX Test do you use the in game FPS test? i was wandering so i can compare to see how my 2x 4850 do
Nicely done W1zz. The temp looks amazing but i still prefer the reference cooler.
Great to see it nearly matches its big B once over-clocked.
Good review, though i have to yet again protest against "No support for CUDA/PhysX".
As i've said in some other review thread, this just can't be considered as con. It's like saying that NVIDIA Riva TNT2 not supoorting 3dfx Glide was considered as con. It just makes no sense. Sure it sucks, but you just can't list it there under cons.
880Mhz for $350 shipped. Not bad, but I still can't believe my $279 shipped (on sale a month ago) nothing-special Sapphire non-ref does 960/1200 on stock volts. I keep waiting for stuff to crash or at least artifact gaming at 950/1150 but it never does. At least so far, not even BC2.
Hey look it's now got a $25 off promo code and free shipping ($289 shipped):
Good review, clearly the card only has minimal performance gains due to the clock speeds. Which can be clocked even higher on a reference gpu. Nice little cooling design. I'll stick with a reference build though for voltage goodness.
Nice review. It seems more and more that the non reference cards are not voltage tweak friendly Must be far cheaper to use the chip they are using now.
Where on earth have you seen a 5870 in the 310 range. Maybe you meant a 5850?
Thank you for including 1280x1024 resolution!! Most review sites do not include this resolution.
I use an HDTV with a native resolution of 1360x768, which would yield about the same benchmark results as 1280x1024.
Some people like reference coolers, I do not, they are much more difficult to clean, and you have to make sure none of the tiny screws get lost or forgotten if you disassemble a reference cooler for cleaning.
I have an HD 4890. The big problem with them at idle is ATI had not figured out how to throttle down the memory speed and that wasted power. CCC does offer manual custom settings and I have memory at 500MHz for desktop use, and it saves quite a bit of power [sorry, I had the exact numbers written down, however, I have been very busy with important matters and cannot remember where I put my notes].
Is it comparable? I mean, 1280 x 1024 to 1360 x 768
1280 x 1024 = 1.310.720 pixels.
1360 x 768 = 1.044.480 pixels.
1.310.720 - 1.044.480 = 266.240
Is it OK to "ignore" those two hundred thousand pixels to have an estimated performance in 1360 x 768 resolution?
Thank you for input.
I have never seen benchmark results done that included 1360x768, and I doubt that benchmarks will ever be provided at that resolution.
The thrust of my comment was that 1280x1024 benchmarks were included in the above review, many reviews from other sites do not include 1280x1024 [leaving low resolution folks in the dark], and again, it is very unlikely that 1360x768 benchmarks will be compiled and provided online.
For whatever game a person plays, given the proper CPU GHz, data throughput, and video card, there will be very little real world differences between benchmarks run at 1280x1024 and 1360x768, except that the latter should run slightly faster if video drivers are optimized.
I have found that my computer does very well when gaming. The only reason I have run benchmarks on my machine is to see how many frames per second I get in the games I play. There are only a couple of games that I have [and currently do not play, they are not even installed] that would drop frame rates below 60. Most of my games bottom out around 100fps or so. I doubt that few people can tell the difference between 60fps and 120fps minimum frame rate.
My statement was not primarily for myself, but, for all those people that use HDTVs as a computer monitor. 1280x1024 will get them in the "ball park" for benchmark results. Yes, there will be a few fps difference, however, the two resolutions are close enough, in my mind, to let people interpolate and determine a fairly good picture as to how computers will perform using screens running at 1360x768 resolution.
Once again, thank you for your comment.
I wondered the same thing.
An off topic note. I have seen many "smiles" on web sites and for the most part most are the same, but this one is an exception
Separate names with a comma.