Printable View
GTX380 Price and Power usage will rival HD5970 (290W~ish.)Quote:
Originally Posted by TurboDiv
g*dd*mm*t.. I hate it when I post news and no-one quotes me within one minute.. you bunch of haters! Are you refreshing Page 48 or too busy editing the tags?
Also, there's a bigger chance of GTS parts arriving to market first. Priced at a much more sane HD5770/HD5830 level.
thought nvidia said it will be a few months till the lower end cards will come?Quote:
Originally Posted by neliz
mid-range parts were always scheduled for Q2, a full half year after the supposed introduction of GF100.Quote:
Originally Posted by Jamesrt2004
Based on GF104 ?Quote:
Originally Posted by neliz
Yes, gt200 replacement, like juniper replaced RV770.Quote:
Originally Posted by Olivon
edit: Okay, who's the joker with the Robocop tag?
good point... :)
any admins see this please sort out the tags its anoying now.
The tags are hilarious :rofl:
loool @Tags just saw that!
Soundz good for a price war :up:Quote:
Originally Posted by neliz
15000P not bad at all ...
fakex0rs :confused: I posted it, might be true, might not be. Just strange that donanimhaber.com posted it just 2 hours later..Quote:
Originally Posted by Olivon
:lol: :rotf:Quote:
global warming
ROFL yeah I especially love the Global Warming and "A wild Fermi appears!" ones.Quote:
Originally Posted by Luka_Aveiro
Humor is bad!!! BAD!!!! :eek: .Quote:
Originally Posted by Jamesrt2004
I loved the
I'd buy that for a dollar
ATI #2 again
global warming
will it play crysis
yes it will @ 100fps
Damn this is killing me.
I think whoever put that second tag has played way too much Pokemon or way too many RPGs :DQuote:
Originally Posted by GoldenTiger
---Edit:---
http://i127.photobucket.com/albums/p.../wildfermi.png
---End edit---
Seriously, though, where's the info? I want to see what Nvidia has (or doesn't have) to offer!
Didn't Nvidia claim they would get the dual GPU cards out about 3 months after the single GPU cards launch?
and passive glasses dont?Quote:
Originally Posted by Teemax
i know, but i dont see how this is a mayor issue that cant be overcome... and i havent noticed any problems with this when using the glasses tbh... i havent used them much and not for several hours in a row though tbh...Quote:
Originally Posted by Teemax
yes, this is what i notice as well and its very annoying... in the crt days everybody used 60hz or 85, while for me there was even a difference between 100 and 120hz... not a big one, but i def liked 120hz more (had a very nice belinea monitor :D)Quote:
Originally Posted by Teemax
i remember when i visited somebody elses place and saw their 60hz screen i seriously couldnt use it... 85hz was acceptable but was annoying.
anyhow, im drifiting ot :D
i think the flickering problem will be solved with 200hz+, giving each eye 100hz, and its probably not as much of a display flicker thing, but more of a periphal vision flicker problem, as our periphal vision is a lot more sensitive to fast movements, especially if its high contrast stuff (bright room/closed shutter)
this is something that might be an actual deal breaker for this tech... while i agree that this might not be the best way to do 3d, i think its def the best out of the possibilities we had to get this stuff started (again)... i mean theres a reason 90% of the display vendors decided to go for the active shutter glass tech...Quote:
Originally Posted by Teemax
yeah i know about all those issues, but every tech has its drawbacks... ive been at the display tradeshow here in taipei last year where the big panel makers showed all their latest stuff, and they had several 3d techs... and the only decent one was with active shutter glasses... :shrug:Quote:
Originally Posted by Teemax
there was one with passive glasses and a double panel merged into one with interleaved horizontal lines to create one matrix with different polarization for each line, AND the cheaper version with a conventional panel with plastic foil on top that changes polarization of half of the horizontal lines. those worked alright, but not great... the active shutter glass stuff looked much better to me...
that video is so not funny to me... :shrug:Quote:
Originally Posted by neliz
GF104 out before GF100? any idea on specs or target perf of GF104? GTX285 perf+dx11?Quote:
Originally Posted by neliz
yeah but they said that a few months ago ^^Quote:
Originally Posted by Jamesrt2004
nah, no way man... nvidia doesnt use definite terms like that when talking about the future... :D they said "real soon" and "only a few months" and stuff like that all the time... the only definite timeframe they used was "november", cause they had to use at least ONE defined timeframe...Quote:
Originally Posted by Carfax
and who knows, maybe they will launch a dualgpu card in november 2010 and then say "we told you it would launch in november..." :shrug: :lol:
^ "fermi in november" tag
:rofl:
30 minutes to more tags..Quote:
Originally Posted by jfromeo
:rofl:
Now is 5:20 PM on Pacific Time (Ex: Los Angeles)... so more 3 hours and 40 minutes.
Lol, it's probably a lot easier to just come back at that time though :)Quote:
Originally Posted by jfromeo
@zalbard, the 5970 SIX will be a good counter if it has higher clocks and performance, not because it has 2GB RAM or six display outputs.
yeah, totally agree...Quote:
Originally Posted by trinibwoy
~3~ hours left if tonight :D
if so keep an eye on NVDA stock ;)
invest.. make $.. get you that fermi!
no one is gonna be awake at 9pm pacific time, so lame
wow this whole time i thought it was saturday. though you guys were crazy.:rofl:
i think the best counter for ATi is not more performance but lower price point. just like HD4000. its quite clear that is the ultimate goal of small die strategy. nvidia puts absolute performance first as it is good for their image. nv needs to get their power consumption higher to compete with ATi's dual gpu strategy. this makes a 250 watt single gpu comparable to a 300 watt dual gpu. its like nehalem EX of gpu's.
That is not going to happen... Nvidia's chip are bigger (has more transistors) so is more expensive to make them and produce less chips per waffle then RV870.Quote:
Originally Posted by Chumbucket843
Unless they want to lose money, lower final price then ATi cards is not going to happen.
And also, i am pretty sure, the day GF100 hot the street, ATI will lower their current prices.
:yepp: :(Quote:
Originally Posted by Manicdan
:rofl:Quote:
Originally Posted by jfromeo
I love the will it play crysis and 000 tags.
I will be since I am in the PST time zone. Hehe, you go ahead and have a good night. :p: :)Quote:
Originally Posted by Manicdan
I will be :D I don't expect much info to actually show up though. I don't even expect real benchmarks.Quote:
Originally Posted by Manicdan
i will be, here at work pressing F5 to pass the time... :DQuote:
Originally Posted by Manicdan
Charlie's own interpretation of the infamous NDA:
http://www.semiaccurate.com/2010/01/...anufacturable/
enjoy the read!
you gotta enjoy charlies articles just for the laughs LOL :DQuote:
Originally Posted by sbattosai
i find him abit crazy...i hope hes not right about fermi tho
I hope he is. 60% increase over the HD5870 is great, and he says that like it's a bad thing :ROTF:Quote:
Originally Posted by jazkat
Yeah because it's 60% in cherry-picked benchmark scenarios. Real life performance should be significantly lower.Quote:
Originally Posted by Redsand426
Anyway, as much as I despise his writing style he does raise some very relevant points. And he has been rather reliable as of late...
lol the tags on this thread are hilarious :rofl:
Yeah, because everything is "cherry picked" when the outcome doesn't give you the results you were hoping for. /sarcasm. :rolleyes:Quote:
Originally Posted by Boissez
Beat me to it :p:Quote:
Originally Posted by SKYMTL
Calling "cherry picked" on benches that aren't even out yet? :ROTF:
Feels like AMDZone in here sometimes.
this is from a nvidia fanboy at DH, you gotta admit it's funny:
FERMI
http://www.donanimhaber.com/image.as...n4_dh_fx57.jpg
ATI 5000
http://img7.imageshack.us/img7/6206/16ive2ujpg.gif
Articles are about to come up... http://www.hardforum.com/showthread.php?t=1487318 HardOCP shows that their article will be up at midnight EST (19 minutes from now).
No I said cherry picked because of what was actually written in Charlies article:Quote:
Originally Posted by SKYMTL
Now - i'm no position to assess whether the lack of hardware tesselation is going to have any performance impact - but *usually* emulation takes more processing cycles than hardcoded funcions.Quote:
When showing off the tessellation capabilities last week, Nvidia was very careful to show the Unigine/Heaven benchmark with the dragon statue, claiming it beat the ATI 5870 by 60%. While technically true, this number is purposely misleading on two accounts. First is that this is a best case for showing off tessellation and and only tessellation. If they were to show off any of the other Heaven tests, the margins would narrow significantly.
15 more minutes !!! Im tired of speculations. Nvidia better be ready or I'm jumping ship... :mad:
His statement is semi-accurate which is what I was referring to in my statement. I'll be able to talk more in 15 minutes. ;)Quote:
Originally Posted by Boissez
It's 5:45am in Spain and I have to go to work in 2 hours :P
looks like the slides did their job, ppl who have them talk about how great Fermi is
His usual spin seems strained this time. I'll bet Fermi is going to end up delivering after all.Quote:
Originally Posted by sbattosai
5 minutes before we find out what we probably already know :toast:
Thanks Golden :)
Anything new in that? Don't feel like reading through stuff that doesn't include actual game benchmarks.Quote:
Originally Posted by GoldenTiger
Nothing much, tessellation performance as previously rumored, but no actual gaming perf. numbers.
What a surprise :eek:Quote:
Originally Posted by annihilat0r
Realy not much there. Its just slides of stuff that's kinda known and slides of what they say the performance will be. No real benchmark test done by reviewers.
The image linked is pretty much all there is in that article other than an off-the-cuff remark mentioning probably 20% better all-around performance than the 5870. There's also a video showing Far Cry 2 in action comparing it to the GTX 280 right off next to eachother on two monitors.Quote:
Originally Posted by kgk
Including architecture scaling, GAMING PERFORMANCE and a whole wack of other stuff.....
RIGHT HERE!!!!!
http://www.hardocp.com/image.html?im...8yNV9sLmdpZg== One of the slides shows a few nv-provided performance #'s.
EDIT:You rock!Quote:
Originally Posted by SKYMTL
EDIT: Wow... great article, and the gaming results are insane...
P.S. 5870 for sale!!! :p:
EDIT2: The HardwareCanucks article is by far the best, most of the others just showed the slide deck and some generic commentary. Yours has actual stuff written out, some educated guesses, and the exclusive performance info :D.
GF100 is getting like Duke Nukem Forever. We keep seeing teasers but never anything useful.
ok im waiting fermi as much as the rest of you but where the hell are the numbers?
it has been shown running benchmarks so why not get the numbers out? there not doing a good job trying to stall ati's sales...
im not sure guys im not liking this secrecy.
edit: pk so we have some number :P
:up:Quote:
Originally Posted by SKYMTL
Thanks for the compliment.Quote:
Originally Posted by GoldenTiger
People here at XS would also do well to take a gander at the Architectural Scaling section for some hints as to what COULD be coming. ;)
As expected: Something in between 5870 and 5970 considering performance...
Ever since I first discovered your site last year, it's really started to grow on me. Solid reviews and well written articles.Quote:
Originally Posted by SKYMTL
Keep it up!
Wow, it looks really promising! I'm glad that I didn't immediately hop on the ATi boat. But let's see if NV can put out a reasonably priced card.
Fermis are here at PDXLAN :)
:) Read through a lot of the article now, comment still holds ;). See edit2 above.Quote:
Originally Posted by SKYMTL
Agree wholeheartedly. Hardware Canucks is by far the best article. :up:Quote:
Originally Posted by GoldenTiger
Looks like they did a very thorough job of extrapolating real world figures based on the live demo.
The minimum FPS on FC2 @2560x1600 is pretty impressive. Where does the 5970 fit on that graph out of curiosity?
CUDA Cores = 512 scalable into 32 core chunksQuote:
Originally Posted by H2O
Texture Units = 64 scalable into 4 unit chunks
Memory: 384-bit scalable into 64-bit partitions
ROPs: 48 scalable into 8-unit groups
Memory interfaces, ROPs and L2 Cache must be scaled together.
Don't know about the other articles but it's in mine...
Here's the section about architecture scaling: LINK
you mean he wasnt reliable in the past? :confused:Quote:
Originally Posted by Boissez
IF fermi really performs as well as a 5970, that will be pretty impressive...
if you think of efficiency, thats way better than cypress... to deliver the same perf as 550mm2 big fermi, ati actually needs two 333mm2 cypress cores, which is 550 vs 666. with perfect yields that is... ;)
even with all the broken part recycling aka 360s im sceptical on how much money nvidia can make with gt300 parts, at least on 40nm... i guess thats not what fermi is meant for though, nvidias cash cow seems to be the 350... i cant wait to hear more about THAT one! :)
hahahah awesome! :DQuote:
Originally Posted by annihilat0r
i wonder how biased the slides are... if nvidia claims too much its going to backfire on them... big time... if you raise the expectations too much, especially with pc enthusiasts, youll run into issues even if you have a great product at a great price :DQuote:
Originally Posted by annihilat0r
So NVIDIA ran the benchmarks for GF100? Just clarifying, as that's what it sounded like.Quote:
Originally Posted by SKYMTL
Not really exclusive. Many people in my session were busy recording it on their cameras. Me, I bolted for a piece of paper to write down the results. :up:Quote:
Originally Posted by GoldenTiger
Ah, well that answers the question :DQuote:
Originally Posted by SKYMTL
Yes but it was the standard Ranch Small test in the Far Cry 2 benchmark. In addition, the program itself wasn't running in the background in any way.Quote:
Originally Posted by randomizer
They booted the benchmark, changed the settings in front of our eyes and then ran it. Simple as that and like anyone else would do. The only difference is that none of us was holding the mouse. Basically, it doesn't make one lick of difference whether it was an NVIDIA system or not; they stated the specs, ran the same tests 99% of the other review sites usually do and showed results. ;)
Me, I replicated their test system as outlined by NVIDIA and popped up the HD5870 result. They weren't BSing either since I was able to replicate their GTX 285 results down to a few percentage points.
Review up @ AT too.
http://www.anandtech.com/video/showdoc.aspx?i=3721
So you have no idea if the bench program has been messed with. I fell better if a independent reviewer could install the OS and bench software then them just handing a system and say just run the bench.Quote:
Originally Posted by SKYMTL
/tinfoilhatQuote:
Originally Posted by SnipingWaste
Thanks :up:Quote:
Originally Posted by SKYMTL
simulated performance on the 5870 in fc2? lulz. Of course they simulate way too low.
I just ran these and 71.66 fps on stock 5870. Overclocked to 1000/1300 5870 and resulted 79.54 fps. Dual 5870's @ stock resulted in 120.35 fps.
As I mentioned in the article, NVIDIA sent me their EXACT system specs when I tried to call them on using a hand picked system. I went out and bought an i7 960 and replicated their system. My GTX 285 results were within a mere few percentage points of theirs.Quote:
Originally Posted by SnipingWaste
Trust me, nothing was messed with the benchmark on my system and judging from my results, nothing was "optimized" on theirs either.....
It was me that simulated the HD 5870 after simulating the GTX 285 results they gave out. Trust me, unless they used a totally different test system for the GTX 285 and GF100, the results are pretty accurate. :DQuote:
Originally Posted by NCspecV81
Meanin you ran on DX9. In DX10, I got nearly exactly the same result using a stock 5870.Quote:
Originally Posted by NCspecV81
you simulated. enough said, they are inaccurate. I'm even running a phenom II to boot.Quote:
Originally Posted by SKYMTL
No it was dx10. I would be more than happy to post the graph.Quote:
Originally Posted by annihilat0r
better 8AA scaling, cool...Quote:
Originally Posted by GoldenTiger
and supposedly better AA quality as well... nice!
very nice perf scaling compared to a 285... but without knowing what res this was its not that useful...
lol from the pcp article:
errr nOooOoO? doom3 faces sucked...Quote:
Remember how impressed we all were when Doom 3 came out? How we now had human looking faces?
soooo wait... i thought the 360 was supposed to be 20% faster than the 5870...Quote:
the GF100 based high end card will be 15% to 20% faster overall than the HD 5870.
so fermi is as fast as a 1ghz 5870? :/
http://img62.imageshack.us/img62/279...lationperf.jpg
fermi is 6x faster in an nvidia custom benchmark compared to a 5870...now thats a useful perf indication! NOT! :P
http://img195.imageshack.us/img195/5875/wptextperf.jpg
250% the logic but only 50% perf boost? or is this some synthetic texturing benchmark? even if its only texturing perf, 50% isnt really a lot... shouldnt a 295 beat fermi in texturing perf then?
meh, this is all?
no perf numbers, no price, no detailed availability/launch date?
nothing about GF104? no pictures of the card? no pictures of the gpu die? no infos about ocing, temperatures in idle and load? :/
then it has got something to do with you running on Phenom. AT was telling that Nvidia + Phenom could perform better than Nvidia + Intel.Quote:
Originally Posted by NCspecV81
Phenom. 'Nuff said. Different mobo, processor architecture, PCI-E lane configuration from the chipset, not DDR3 tri channel....need I go on?Quote:
Originally Posted by NCspecV81
Honestly, these results are easily replicated with the proper system (960 or i7 920 overclocked to a 960's Turbo Mode speed) and running the same benchmark with 9.12 or 9.11 drivers.
I've never been able to achieve the kind of performance you're claiming... and I'm running an i7 920 @ 3.72g, an Indilinx Barefoot 256GB SSD with Samsung NAND, 6GB DDR3 @ 1560, with a Radeon 5870 (stock).Quote:
Originally Posted by NCspecV81
Don't tell me when it comes to FarCry2, Intel sucks compared to AMD.
Yea Far Cry 2 doesn't seem to give Core i7 much love, or it gives extra love to Phenom II. :DQuote:
Originally Posted by annihilat0r
cool, thx for spending so much time on this man! appreciated :toast:Quote:
Originally Posted by SKYMTL
so best case scenario in a tesselation benchmark, which to some degree already is a best case scenario, fermi is 50% faster than a 5870... hmmm so its slightly faster than a 5970 in heavy tesselation scenarios.Quote:
The cart you see below Unigine engine benchmark 60 seconds is taken from the walkway section of Heaven benchmark that has the most tessellation.
typo, theres a word missing after seperate :DQuote:
NVIDIA showed off one more bit of information about DX11 performance which was from a Microsoft DX11 geometry shader toolkit. The results you see below were taken during the rendering of two separate which ask the GPU to process up to six cubemap faces in one pass.
so was this a microsoft dx11 benchmark? or is it an nvidia benchmark based on microsofts toolkit? :D
looks like I expected it to
sorry turned down my OC to test nearly the same frequency. If you are trying to tell me a phenom performs much better in fc2 than intel...give me a break. There's been plenty of reviews that state otherwise.
http://www.overclock.net/gallery/dat...gle5870fc2.PNG
SKYML:
great article. Best article out there, starting to like your site over AT. keep up the great work.
Image thumbnail, please!
Sorry. A lot of stuff to type.Quote:
Originally Posted by saaya
The benchmark is built into MS's DX11 toolkit for testing of geometry shader code. It is an industry standard with developers as far as I know. I have asked for a bit more information about it but my developer contacts don't think it is publicly available.
Stay tuned for more on that.
It's a Canadian thing. :p:Quote:
Originally Posted by kgk
Um.....what? :confused:Quote:
Originally Posted by annihilat0r
I don't know which article, but it was there.Quote:
Originally Posted by Carfax
that's what I'm saying. The results posted for the 5870 are "simulated" and a mild guess at best and are "based off the performance obtained with the gtx285". Great. Just what we need floating around, especially for cards already out on the market!Quote:
Originally Posted by Carfax
overclocked results lol
pardon me I don't know how to put in thumbnail tags. Let me know and I will.
http://www.overclock.net/gallery/dat...fc2results.PNG
God, why on Earth do you continue posting overclocked results while it has absolutely no relevance?
Here is something from AT, with a 3.33GHz i7:
http://images.anandtech.com/graphs/r...2344/20103.png