Van Owen said:Which is why he said it was fake. Why are you calling me illiterate?
Elixist said:Wtf? I smell bullshit. Link please I would love to see that.
How about people wake the fuck up and realize no system launching in 2012 or 2013 is going to be using 2012/2013 technology.Van Owen said:Again, the system isn't hitting until mid 2012. How excited can everyone be expected to be by 2008 tech?
IchigoSharingan said:http://slickdeals.net/forums/showthread.php?sduid=913724&t=2924373
http://slickdeals.net/forums/showthread.php?t=2925269
http://www.google.com/products/cata...WtQP4j9W-Cw&ved=0CGMQ8wIwAA&biw=1920&bih=976#
http://www.newegg.com/Product/Produ...-Memory+(Desktop+Memory)-_-Corsair-_-20233132
i5 2500k + 1TB HD = $200
Radeon 5830 = $100
mobo =$111
ddr3 1333 ram =$40
power supply/dvd writer easy to find deals on.
Look a few posts up I linked it.Elixist said:Thats not wat he posted, he said a prebuilt pc. Look at the manufacturers of prebuilts and lol at what you get for 500 bones.
Elixist said:Thats not wat he posted, he said a prebuilt pc. Look at the manufacturers of prebuilts and lol at what you get for 500 bones.
The screens are not small. They're rumored to be in the 800x500 range (some time ago I suggested them to be 960x540 so that 4 of them would be exactly 1080p), and rendering an additional nearly-1080p screen always split in four is going to eat some significant resources.IchigoSharingan said:The screens are small though. That would offset the intensive nature of the stream on the CPU. 4 streams at once... maybe one core could handle all of that. But the rumor is triple core, right? 2 cores left for the actual game video + sound.... hmm...need a 4th core.
We need to know:
quality
data rate kbit/s
video only?
codec
frame rate
Someone will still argue they can do it by using NANOMACHINESantonz said:How about people wake the fuck up and realize no system launching in 2012 or 2013 is going to be using 2012/2013 technology.
Elixist said:thats more like 700 and somethin dollars dude after conversion, a bit different. still a good deal tho i built a similar config in newegg for about that same price.
back on topic cant wait for the eeee threees
Jocchan said:The screens are not small. They're rumored to be in the 800x500 range (some time ago I suggested them to be 960x540 so that 4 of them would be exactly 1080p), and rendering an additional nearly-1080p screen always split in four is going to eat some significant resources.
SneakyStephan said:I wish conversion worked that way for hardware and software in the EU.
They replace the dollar sign with a euro sign and call it a day.
Console games cost 60-69 euros, steam games cost 49 euros (some 59 like ubisoft and activision shit), ps3 on release was 599 euros..., gtx580 is 530-550 euros , sammy spinpoint f3 hdds are 65 euros etc etc.
Items that get a proper conversion rate are few and far between, and usually only due to sales or overstock.
Nintendo wouldn't release a system heavily unbalanced or fatally crippled by easily noticeable bottlenecks.IchigoSharingan said:of course, but Nintendo has put tons into R&D, so it's safe to assume they have addressed this issue
szaromir said:Radeon 4850 would put it in 5 times more powerful than 360 ballpark. Hell yes if true.
SneakyStephan said:And here's what the bandwidth would look like with gddr5 instead of ancient gddr3. KaPOW.
SneakyStephan said:Btw my phenom II 720BE is (only) clocked at 3.2 ghz atm to keep down fan noise, so I'm closer to the specs.
Jocchan said:The screens are not small. They're rumored to be in the 800x500 range (some time ago I suggested them to be 960x540 so that 4 of them would be exactly 1080p), and rendering an additional nearly-1080p screen always split in four is going to eat some significant resources.
Well yeah, because the fillrate isn't much better and the core clock isn't much higher.DeFiBkIlLeR said:But that wont deliver twice the performance desipte it showing 115gb compared to my 60gb, the difference in performance between a 4850 and a 4870 is roughly 20%, and 40% compared to a 4890.
Lets split the difference.. seeing as the leaks say its a higher clocked version of the Xbox CPU...)
Efficiency (GFLOPS/ Watt) is almost exactly the same for all 40nm GDDR5 GPUs.Mr_Brit said:Both the 5000 and 6000 series from AMD are quite a bit more efficient than their predececssor even though they're all built on a 40nm process and use basically the same achitecture. By the time Sony/MS new systems are out we'll have new architectures from both AMD and nvidia and they'll be a lot more efficient as they won't be adding new features like the previous generations but will focus on pure power.
antonz said:Pixel Fillrate these days is a useless figure for determining performance. It hasn't been a determining factor of Card performance since around 2003.
The Black Brad Pitt said:Can you elaborate please?
How does it stack up next to a 4870? 5870?
Pixel fill rate will be about 2.5x Xenos capability. Xenos does 4 gigapixels a second.Graphics Horse said:It sure would be nice to have enough fillrate to pull off nice alpha effects though.
http://www.s314kba.co.uk/images/lukesr/GT5/Autumn Ring_7.jpg
Graphics Horse said:It sure would be nice to have enough fillrate to pull off nice alpha effects though.
http://www.s314kba.co.uk/images/lukesr/GT5/Autumn Ring_7.jpg
Risk Breaker said:1GB RAM, 3.2ghz tricore ibm aka 360.0 and an 4800... Absolutely won't buy. By the time the hardware is used properly, the other two will be out and will be significantly better. Pass.
That's pre-patch. Not nearly that bad now.
Frankly im getting to the point id give up my first child just to have Sony and Microsoft both cheap out next gen just to see the meltdowns.Uchip said:buying a console based solely on the comparison to the competitors specs seems really irrational
Risk Breaker said:1GB RAM, 3.2ghz tricore ibm aka 360.0 and an 4800... Absolutely won't buy. By the time the hardware is used properly, the other two will be out and will be significantly better. Pass.
Really? The 570 is pretty much as good as a 480 but uses vastly less power and is built on the same process. The same also applies to the 6870 and the 5870.wsippel said:Efficiency (GFLOPS/ Watt) is almost exactly the same for all 40nm GDDR5 GPUs.
Yep (see my other post). Many games will probably use the controller screens for trivial stuff, perhaps even just 2D menus or maps.Graphics Horse said:Of course, but 3DS games haven't tended to put significant resources into the bottom screen so far. The opportunity is there of course, but I don't think games rendering 5 different 3D viewpoints at once will be the norm.
It might be that a 4 player game would force you to reserve an extra 4.5MB of frame buffer RAM no matter what you chose to show, limiting the image quality of the TV screen, but it's too early to say.
Risk Breaker said:1GB RAM, 3.2ghz tricore ibm aka 360.0 and an 4800... Absolutely won't buy. By the time the hardware is used properly, the other two will be out and will be significantly better. Pass.
Probably closer to ~150W, maybe ~180W if IGN is correct regarding the size of the system.Hazaro said:Also keep in mind the power draw under load is going to be like 100W compared to a loaded PC with those specs running at 250W.
The rumoured specs suggests the system won't get anywhere near 180W, what would give you that idea?wsippel said:Probably closer to ~150W, maybe ~180W if IGN is correct regarding the size of the system.
Nope.Mr_Brit said:Really? The 570 is pretty much as good as a 480 but uses vastly less power and is built on the same process. The same also applies to the 6870 and the 5870.
I was about to post this but you beat me to it.Mr_Brit said:Really? The 570 is pretty much as good as a 480 but uses vastly less power and is built on the same process. The same also applies to the 6870 and the 5870.
That's a terrible comparison, both 480 and 570 are a lot faster than the 5870 yet have lower TFLOPS. Also, the 6870 uses less power than the 5870 so I don't know where you got these numbers from. Also, the 5000+ series are DX11 compliant so need a lot more transistors to support those features which is why their TDP is higher than you'd expect.wsippel said:Nope.
5870: 2TFLOPS @ 151W
6870: 2TFLOPS @ 151W
480: 1.5TFLOPS @ 250W
570: 1.4TFLOPS @ 219W
And just for shits and giggles:
4770: 960GFLOPS @ 80W
5770: 1TFLOPS @ 86W
6750: 1TFLOPS @ 86W
Nvidia managed to improve efficiency a bit, but still looks pretty bad compared to AMDs offerings in that regard.
You don't really need a case the size of an original Xbox360 unless you expect a significant power draw. Especially not if you're Nintendo and well known for designing small, efficient hardware.Mr_Brit said:The rumoured specs suggests the system won't get anywhere near 180W, what would give you that idea?
Uchip said:buying a console based solely on the comparison to the competitors specs seems really irrational
neoanarch said:You're talking about something 3-4 years in the future. Plus I wouldn't bank on Sony/MS breaking the bank one more time. They won't be 600 dollar machines when they launch.
It's the only sane comparison, because it completely excludes code and driver level fuckups or one sided optimizations which are way to common on the PC ("The Way it's Meant to be Played"). Also, the 4770 achieves 12GFLOPS/W, the 5770 11.7GFLOPS/W. The 5770 is less efficient.Mr_Brit said:That's a terrible comparison, both 480 and 570 are a lot faster than the 5870 yet have lower TFLOPS. Also, the 6870 uses less power than the 5870 so I don't know where you got these numbers from. Also, the 5000+ series are DX11 compliant so need a lot more transistors to support those features which is why their TDP is higher than you'd expect.
Edit: Going by your chart, the 5770 uses 6W more but is significantly faster than the 4770 which proves my point.
Terrible comparison, use cards that have similar power instead of cherry picking examples. Using your example I could say that a 5200 card is more efficient than a 6900 card which would be strectching the truth even if it is true. Take the 6870 and 5870, nearly the same power but the 6870 uses quite a bit less power.wsippel said:The 4770 achieves 12GFLOPS/W, the 5770 11.7GFLOPS/W. The 5770 is less efficient.
80W and 86W are as similar as it gets. I did compare cards with similar power levels. Also, we're talking about consoles here. Driver optimizations or code optimizations have no place in this discussion.Mr_Brit said:Terrible comparison, use cards that have similar power instead of cherry picking examples. Using your example I could say that a 5200 card is more efficient than a 6900 card which would be strectching the truth even if it is true. Take the 6870 and 5870, nearly the same power but the 6870 uses quite a bit less power.
Who says their gains are anything to do with improved drivers? 5870 and 6870 are practically identical, drivers wouldn't be able to substantially improve the 6870 and not do the exact same for the 5870. Also a 4770 and 5770 aren't the same power, as you increase in power, your efficiency always goes down which is why you should only compare cards that are very close to each other in performance.wsippel said:80W and 86W are as similar as it gets. I did compare cards with similar power levels. Also, we're talking about consoles here. Driver optimizations or code optimizations have no place in this discussion.