R700 Launch in for Wet-Weather, GTX 280 Slash Coming Up

[farlex85]

meh i cant see 80% lead myself seeing as how a single card preforms compared to a GTX280 its damn near impssible without 100% efficny whish is IMO impossible.

Well, in some benchmarks they actually have achieved very near 100% efficiency on some of the latest cards from ati and nvidia. Ati also is saying they have a new way of bridging the gpu's to avoid problems that normally arise w/ cf. Still, I would think 80% average is pushing it too, more like 50-60% maybe. We shall see though.

 

[candle_86]

yea honestly, and 100% isnt possible, you loose data simply be transfer, no GPU is 100% effecint, nothing can ever be, thats a rule of pshycis

 

[Megasty]

meh i cant see 80% lead myself seeing as how a single card preforms compared to a GTX280 its damn near impssible without 100% efficny whish is IMO impossible.

You would be surprised to see how close those 2 cards are if you had the cards firsthand. Benchmarks are garbage. They have their biases all across the board.
Of course sometime the cards are duds but when you get a great versions of binned GPUs, you can tell right off.
There are games where the GTX280 is 25% faster than the 4870 then there are games where that closes to 10%.
But that still means that the 4870x2 will have to have 100% eff. in order to do that 80%.
Games that can take such performance for the 4870x2 already exist & its clear to which ones they are because they can already do 200+ fps maxed out with CFx.

 

[candle_86]

You would be surprised to see how close those 2 cards are if you had the cards firsthand. Benchmarks are garbage. They have their biases all across the board.
Of course sometime the cards are duds but when you get a great versions of binned GPUs, you can tell right off.
There are games where the GTX280 is 25% faster than the 4870 then there are games where that closes to 10%.
But that still means that the 4870x2 will have to have 100% eff. in order to do that 80%.
Games that can take such performance for the 4870x2 already exist & its clear to which ones they are because they can already do 200+ fps maxed out with CFx.

but none the less 100% is impossible for any device, phsyics apply to electronics, we loose efficny due to heat, travel time, ect. 90% i can belive possible

 

[farlex85]

You would be surprised to see how close those 2 cards are if you had the cards firsthand. Benchmarks are garbage. They have their biases all across the board.
Of course sometime the cards are duds but when you get a great versions of binned GPUs, you can tell right off.
There are games where the GTX280 is 25% faster than the 4870 then there are games where that closes to 10%.
But that still means that the 4870x2 will have to have 100% eff. in order to do that 80%.
Games that can take such performance for the 4870x2 already exist & its clear to which ones they are because they can already do 200+ fps maxed out with CFx.

Game fps are just other benchmarks. Real world performance is somewhat qualitative, thus the need for benchies to quantify it.

but none the less 100% is impossible for any device, phsyics apply to electronics, we loose efficny due to heat, travel time, ect. 90% i can belive possible

But the physics are the same for 1 card as for 2 (essientially). The amount of energy lost as heat and friction and the like does not factor in when talking about scaling across 2 cards, b/c it is already factored in to the overall performance of a single card.

 

[Megasty]

but none the less 100% is impossible for any device, phsyics apply to electronics, we loose efficny due to heat, travel time, ect. 90% i can belive possible

Yeah, that also means that none of these GPU are 100% eff either. So if the 4870x2 has twice the performance of the 4870, its still not 100% eff over the 2 cores. The 4870x2 doesn't have to be 100% eff to be 80% faster than the GTX280 in 2 or 3 games. OK, I need to stop splitting hairs

Game fps are just other benchmarks. Real world performance is somewhat qualitative, thus the need for benchies to quantify it.

I know, just how fraps take away from performance too We don't know how much it really is taking away but atleast its taking it away in all cases

 

[candle_86]

Game fps are just other benchmarks. Real world performance is somewhat qualitative, thus the need for benchies to quantify it.



But the physics are the same for 1 card as for 2 (essientially). The amount of energy lost as heat and friction and the like does not factor in when talking about scaling across 2 cards, b/c it is already factored in to the overall performance of a single card.

true, but a single GPU will always be the more efficent way to go untill game designers make there grahpics engine split to take advantage of multiple GPU's.

 

[PCpraiser100]

true, but a single GPU will always be the more efficent way to go untill game designers make there grahpics engine split to take advantage of multiple GPU's.

Yes, which is why dual core GPUs are out. They basically shave off about 50-75 watts over a dual video card setup, but has very similar performance compared to Crossfire and in some cases SLI. SLI performed so well on the 8000 series that it out values the 8800 Ultra by nearly a $500, which was one of the ony times I started liking NVIDIA. But today's newest CrossfireX has a lot faster and more reliable bandwidth than SLI, so the HD 4870 X2 really pays for itself as it is more affordable than two HD 4870s by a $100 difference. Not much difference right now, but the gap will get bigger in the future.

 

[vsary6968]

Well I just read a preview of tests between the GTX280 and the 4870x2 over at Xtreme systems and the R700 lost about a third of the benches (all were at 16xx and above) but won the two thirds by a fair margin but across the board was 20-25% faster but in some benches 40% faster, of course those were at stock speeds so it does seem that at stock the R700 would even beat an Extreme, of course previews are great but not always the complete answer, for the overclocker though, the decision may be more interesting but I cant help but think ATi will get the pricing more realistic.

You can see that even the GTX 280 SLI is competive to the R700.Go to this website and check out the benchmark. http://www.techreport.com/articles.x/15105 and http://www.anandtech.com/video/showdoc.aspx?i=3354.

 

[farlex85]

You can see that even the GTX 280 SLI is competive to the R700.Go to this website and check out the benchmark. http://www.techreport.com/articles.x/15105 and http://www.anandtech.com/video/showdoc.aspx?i=3354.

Yes but it also costs $880 w/ these price drops for gtx 280 sli. To have it merely compete w/ a $500 card is laughable at best.

 

[PCpraiser100]

Yes, I know the $550 GTX 280 kicks ass at SLI but what will happen if the more kick ass $500 HD 4870 X2 took Xfire for a test drive? I know there are cheaper versions of the GTX 280 but I'm presenting the most clocked GTX 280 from ASUS which is about $550. Approximate MSRP of HD 4870 X2 is $500 so I added that in.

 

[NinkobEi]

i'm curious what crossfired R700s will do...true, thats $1000 in video cards...but they gotta have some serious oomph.

 

[vsary6968]

You know, there are reasons that the 280 beats the 4870x2 in some benches. The main reason is that the drivers are what allows the 4870x2 to copy data over both GPU sets.
When the game doesn't support that then it will behave like one 4870. In that case, the benches show that the 4870 & 4870x2 perform relatively the same.
All the newer game that came out over the past 6 months or so support both SLI & CF.
You won't have that problem now unless the game designers start having massive hangovers on the job.
Also, the 4870x2 is a 2gb card (2x1gb). Some of the ES's were 1gb (1x512mb) which were either a fluke or AMD was trying to test that version of the card as well.
But the launch card will be 2gb of GDDR5. AMD is also going to make a 'super' RV770XT chip that can clock over 1GHz with proper cooling.
We might see a latter version of the 4870x2 that uses 2 of those chips.

All the review and benchmark which I went through all of those site it is a 1gb (512 x 2). The 2GB version haven't sample yet. But this 1GB card still alot faster than GTX 280 already.And also this is only the 1GB version.Their is no 2GB version sample yet. If you can find out if it is 2GB version let me know.If you go to Hexus.net they will say this is a 1GB.

We cannot emphasise enough that the AMD R700, which HEXUS has tested, is an early engineering sample.

AMD has not confirmed the final AMD R700 specification or disclosed final pricing to HEXUS.

The final specifications of AMD R700 may ultimately differ in detail from those we've reported.

Some of the AMD R700 specifications detailed below are a logical extrapolation of known data relating to ATI Radeon HD 4870 in CrossFire.

Furthermore, all measured performance is very likely to improve as the product, its BIOS and its device drivers are optimised to production status, the SKU launches and becomes available to purchase

 

[Megasty]

All the review and benchmark which I went through all of those site it is a 1gb (512 x 2). The 2GB version haven't sample yet. But this 1GB card still alot faster than GTX 280 already.And also this is only the 1GB version.Their is no 2GB version sample yet. If you can find out if it is 2GB version let me know.If you go to Hexus.net they will say this is a 1GB.

We cannot emphasise enough that the AMD R700, which HEXUS has tested, is an early engineering sample.

AMD has not confirmed the final AMD R700 specification or disclosed final pricing to HEXUS.

The final specifications of AMD R700 may ultimately differ in detail from those we've reported.

Some of the AMD R700 specifications detailed below are a logical extrapolation of known data relating to ATI Radeon HD 4870 in CrossFire.

Furthermore, all measured performance is very likely to improve as the product, its BIOS and its device drivers are optimised to production status, the SKU launches and becomes available to purchase

So DH is BS, I don't think so.

http://www.driverheaven.net/reviews.php?reviewid=588

 

[vsary6968]

If you don't think so , u have to wait and see and I don't want to argue something have not release yet.

 

[candle_86]

the BS Sites include but are not limited to

The Inquirer
Toms Hardware
Rage3d
NvNews
The Registar
Hard OCP

other than that most sites tend to agree on things

 

[btarunr]

I just think this type of posting isn't reporting news, rather it's your opinion on what could potentially occur in the very near future.

Shouldn't this be in an opinion forum, or a hardware forum for that matter? I just don't see why you didn't do that? Possibly because you wanted everyone to see your opinion

The only thing that should have been put in is article for me to actually consider it news is you posting the information on the price drops, and only that news. What you really did was post some links to the price drops (not an official announcement link) and then throw in your opinion on what could happen to the market from a direct result of these price drops.

Great price drops though, maybe now people will feel more tempted to by graphics cards in a time where people are reluctant to buy much of anything besides food and gas.

EDIT: Changed "is" to "isn't".

Flip thru your newspaper, get amazed at the number of news items which become 'merely opinions' according to your description. After quoting facts and links to facts, it is news. Newegg sells a whole fleet of stock-clocked cards for $440, it is news. GPU Cafe 'opionates' later in the thread of a further slash, it's something to wait for, but since $440 GTX 280 is a reality, it's not merely opinion. Wake up.

 

[Megasty]

the BS Sites include but are not limited to

The Inquirer
Toms Hardware
Rage3d
NvNews
The Registar
Hard OCP

other than that most sites tend to agree on things

I really don't want to say it but the last one is BS itself. It was fashioned, formed, & derived from BS. It is the original BS, all others merely follow suit.

Sites that don't try to find out anything about what they're testing also should be included.

 

[From_Nowhere]

Well that's good, GTX 280 is within my reach now. Gotta love competition (even if it is staged? )


Oh and for those looking at sub-par tech reviews and news websites look no further than these:

Sub-par Tech Reviews:
Anandtech
Tom's Hardware
CNET Reviews

Sub-par Tech News:
N4G
The Inquirer
CNET News

 

[candle_86]

anyone remember when tomshardware was actully a good place to get info, what happened, when did there stuff go to crap?

 

[Bundy]

Flip thru your newspaper, get amazed at the number of news items which become 'merely opinions' according to your description. After quoting facts and links to facts, it is news. Newegg sells a whole fleet of stock-clocked cards for $440, it is news. GPU Cafe 'opionates' later in the thread of a further slash, it's something to wait for, but since $440 GTX 280 is a reality, it's not merely opinion. Wake up.

Last time I flipped through a newspaper, the opinions were firmly anchored in the opinions section. Most quality news organisations and reporters strive to improve their reputation by being as impartial as possible. In the end, this determines the nature of the site/paper and its reputation, something I think you could consider a little more before posting.

bta, its great that you are posting up such a huge volume of up to date info and maybe I'm just showing my true bias but I'd just prefer to see less AMD/ATI fanboi flavour in your news. If techpowerup wishes to be known as the place where Nvidia gets trashed and ATI is god, you are welcome to ignore my comments as being ignorant.

 

[btarunr]

Last time I flipped through a newspaper, the opinions were firmly anchored in the opinions section. Most quality news organisations and reporters strive to improve their reputation by being as impartial as possible. In the end, this determines the nature of the site/paper and its reputation, something I think you could consider a little more before posting.

bta, its great that you are posting up such a huge volume of up to date info and maybe I'm just showing my true bias but I'd just prefer to see less AMD/ATI fanboi flavour in your news. If techpowerup wishes to be known as the place where Nvidia gets trashed and ATI is god, you are welcome to ignore my comments as being ignorant.

I'm not biased. I use NV hardware, I'm happier than anyone that GTX 280 sells for $440 (as low as $419 AR), so I could buy one soon. NV isn't trashed in the news, pray where? ATI isn't looked up like God? At least I don't. If the same article was made by say another publication and not me, I bet many wouldn't have said "oh it's FUD, on it's opinion". After putting up facts you still put a cold shoulder to it, It becomes a very bad day for me at office. So much effort in vain.

 

[candle_86]

Last time I flipped through a newspaper, the opinions were firmly anchored in the opinions section. Most quality news organisations and reporters strive to improve their reputation by being as impartial as possible. In the end, this determines the nature of the site/paper and its reputation, something I think you could consider a little more before posting.

bta, its great that you are posting up such a huge volume of up to date info and maybe I'm just showing my true bias but I'd just prefer to see less AMD/ATI fanboi flavour in your news. If techpowerup wishes to be known as the place where Nvidia gets trashed and ATI is god, you are welcome to ignore my comments as being ignorant.

you cant be serious here can you, BTA gets called the biggest Nvidiot around here by the newbs running around, when did he become a fanATIc

 

[btarunr]

That's the problem, Candle. Every time I say I like this card and it happens to be NVIDIA, people call me NVIDIOT, otherwise ATI-biased. Why don't you just call me a 'consumer'? That I'm a person with value-for-money?

 

[candle_86]

lol bta i never call you either, im just laughing about all of this