|
repiv posted:What is the market for this thing in TYOOL 2015, where you can fit a Titan X in a 10 litre case with length to spare?
|
#
?
Aug 27, 2015 11:40
|
|
|
|
| # ? May 1, 2024 02:59 |
|
|
repiv posted:What is the market for this thing in TYOOL 2015, where you can fit a Titan X in a 10 litre case with length to spare? Retardedly overpriced mini-form-factor overheating-prone "console killers" from Alienware and co.
|
|
#
?
Aug 27, 2015 11:44
|
|
|
repiv posted:What is the market for this thing in TYOOL 2015, where you can fit a Titan X in a 10 litre case with length to spare? That's a weird thing to complain about. The form factor is great, and I have a feeling if NVIDIA did it, nobody would be saying "this card is too small". It's the price they've hosed up, everyone will buy 980 Tis (that they can actually find to buy!) instead. vv Well yeah, I wasn't comparing the pricing strategy, just the form factor HalloKitty fucked around with this message at 13:59 on Aug 27, 2015 |
|
#
?
Aug 27, 2015 13:41
|
|
|
HalloKitty posted:That's a weird thing to complain about. The form factor is great, and I have a feeling if NVIDIA did it, nobody would be saying "this card is too small". People would complain if it were priced at $650 though. Everybody was pretty stoked if it was $475. Even I was and I have no use for it
|
|
#
?
Aug 27, 2015 13:51
|
|
|
The Nano could have been the saving grace for this whole release, sitting at 400-450$. A 650$, no one is going to loving buy it, what in the hell is it offering over a Fury X at nearly the same price?
|
|
#
?
Aug 27, 2015 14:01
|
|
|
FaustianQ posted:The Nano could have been the saving grace for this whole release, sitting at 400-450$. A 650$, no one is going to loving buy it, what in the hell is it offering over a Fury X at nearly the same price?
|
|
#
?
Aug 27, 2015 14:11
|
|
HalloKitty posted:That's a weird thing to complain about. The form factor is great, and I have a feeling if NVIDIA did it, nobody would be saying "this card is too small". No one is complaining about the size.
|
|
|
#
?
Aug 27, 2015 14:16
|
|
|
FaustianQ posted:The Nano could have been the saving grace for this whole release, sitting at 400-450$. A 650$, no one is going to loving buy it, what in the hell is it offering over a Fury X at nearly the same price? Yeah, echoing this - the fact that some of their benches show the Fury Nano in spitting distance of (or better than) the Fury X is alarming, especially with the X requiring a CLC and the Nano being completely air-cooled. Could have at least dropped $100 off the Nano just for cooling alone. BOOTY-ADE fucked around with this message at 15:03 on Aug 27, 2015 |
|
#
?
Aug 27, 2015 14:35
|
|
|
DrDork posted:Probably this, especially considering the enormous PR poo poo-storm that was the initial 290X launch with "clocks up to X" that no one could ever hit for more than 15 seconds because lol AMD stock coolers. The Nano is power limited, AMD's estimate is that actual clock speeds will be in the 900 MHz range, the 1000 MHz clock is just so only the power budget will limit it. It's a really niche card, but it looks like they have yields that one or two niches could saturate.
|
|
#
?
Aug 27, 2015 14:35
|
|
|
The target segment for the Fury Nano is what, enthusiast gamers with big wallets, who also want to run mITX, who didn't get a case with provisions to fit a full sized GPU? This is why Nvidia has 82% of the discrete market now. AMD's business plan for this generation has been only to release new 550+ dollar flagship cards, an already tiny market segment, while rebadging the rest of their lineup. Then they gently caress that up by making a GPU so hard to make that AMD and its partners can't actually produce any. Turns out no one wants twice rebadged GCN 1.0 Pitcairns, warmed over Tongas and Hawaii chips you doubled the memory on and added an extra 100 bucks to the price in the market segments people actually buy GPU's in. And you can't buy their high end GPU's. But hey, at least their DX12 drivers are good, for all 10 people who will play DX12 games on an AMD GPU.
|
|
#
?
Aug 27, 2015 14:37
|
|
|
AVeryLargeRadish posted:No one is complaining about the size. Giggidy.  I actually got my old drive cleaned up enough last weekend that I'm going to do a clean install of 10 so I can compare scores between 8.1 and 10. If I get really ambitious (maybe over the holiday weekend) I might even install 7 on a spare drive and be able to compare all 3.
|
|
#
?
Aug 27, 2015 14:52
|
|
|
veedubfreak posted:Giggidy. Sounds like an exciting holiday
|
|
#
?
Aug 27, 2015 14:58
|
|
|
Beautiful Ninja posted:But hey, at least their DX12 drivers are good, for all 10 people who will play DX12 games on an AMD GPU.
|
|
#
?
Aug 27, 2015 15:19
|
|
|
Don Lapre posted:Sounds like an exciting holiday
|
|
#
?
Aug 27, 2015 15:21
|
|
|
Its crazy how much we talk about firestrike when there are these "games" that people play with their GPUs... I think "games" is the right word for them?
|
|
#
?
Aug 27, 2015 15:22
|
|
|
Captain Yossarian posted:Its crazy how much we talk about firestrike when there are these "games" that people play with their GPUs... I think "games" is the right word for them?
|
|
#
?
Aug 27, 2015 15:25
|
|
|
veedubfreak, have you spent more time in firestrike than playing through the Witcher series yet?
|
|
#
?
Aug 27, 2015 15:36
|
|
|
fence hopper posted:veedubfreak, have you spent more time in firestrike than playing through the Witcher series yet? Nah, I've only run firestrike like 5? times. Oddly enough firestrike will run at 1504, but MW:O will crash the driver. Finished Witcher 1, and about 10 hours in to Witcher 2. Firestrike is good for when I'm cleaning up around the house and want to push my clocks. But I think I've gone as far as I can. I actually set my GPU back to defaults last night because it maxes everything out at default clocks and there's no chance of crashing the game 
|
|
#
?
Aug 27, 2015 15:44
|
|
|
I had a 1080p monitor and was content with a 970. I started following this thread awhile ago and now I have a overclocked 980ti and 1440p monitor. Thanks? Also it seems like I'm one of the few people here that didn't run any benchmarks yet. I think I'll fix that this weekend. These games better have achievements.
|
|
#
?
Aug 27, 2015 15:55
|
|
|
Anime Schoolgirl posted:What games, do you mean stuff like DOTA which I play on an Intel HD Graphics 4600? That... Sounds like a game?! Honestly I'm just busting people's balls here, but about 100% of the discussion is centered around synthetic benchmarks instead of games. Makes me feel like I'm in Guru3d or something 
|
|
#
?
Aug 27, 2015 16:05
|
|
|
Captain Yossarian posted:That... Sounds like a game?! Honestly I'm just busting people's balls here, but about 100% of the discussion is centered around synthetic benchmarks instead of games. Makes me feel like I'm in Guru3d or something Finally someone said it - buy your fancy GFX card, overclock it if you must, but for god's sake play some games instead of constantly benchmarking them and returning them because the internet says you should get a higher score.
|
|
#
?
Aug 27, 2015 16:12
|
|
|
I'm going to run some benchmarks to talk about those instead of games. So I don't have to tell people I traded up for a 980ti and have mostly been playing pillars of eternity... It's runs great though! e: I've been playing the witcher 3 also. I will say that upgrading to a 1440p monitor has been one of the best computer related purchases I've made in awhile. Being able to max everything out is great. I had a 6870 before. dangling pointer fucked around with this message at 16:19 on Aug 27, 2015 |
|
#
?
Aug 27, 2015 16:13
|
|
|
Ozz81 posted:Yeah, echoing this - the fact that some of their benches show the Fury Nano in spitting distance of (or better than) the Fury X is alarming, especially with the X requiring a CLC and the Nano being completely air-cooled. Could have at least dropped $100 off the Nano just for cooling alone. I think our perspective has been reversed this entire time on the Nano/X relationship - Fury X is the dumping ground for lovely Fiji chips, considering the performance/watt gap is enormously in favor of the Nano despite being the same chip. It's why they need the CLC, it's why they came first. Someone needs to buy a Nano and slap a solid Liquid loop on it and see where they can take it, because if a CLC Nano outperforms an X then it all but confirms AMD marketing is run in a separate universe where it's normal to have your head shoved up your rear end.
|
|
#
?
Aug 27, 2015 16:24
|
|
|
FaustianQ posted:I think our perspective has been reversed this entire time on the Nano/X relationship - Fury X is the dumping ground for lovely Fiji chips, considering the performance/watt gap is enormously in favor of the Nano despite being the same chip. It's why they need the CLC, it's why they came first. Someone needs to buy a Nano and slap a solid Liquid loop on it and see where they can take it, because if a CLC Nano outperforms an X then it all but confirms AMD marketing is run in a separate universe where it's normal to have your head shoved up your rear end. Well my understanding is that the Fury Nano will be power locked, so you probably can't even overclock it to consistently outperform an X, unless there are provisions to unlock the power/volt/temp limits on the card. The Fury X was volt locked after all. And yeah, this seems like an extremely dumb thing to do on AMD's side of things. Take your best performing chips, power lock them and have it barely outperform the 390X. Take the lovely Fiji's, slap on a CLC and clock them to their limits and sell those as your top of the line GPU. What kind of logic is this?
|
|
#
?
Aug 27, 2015 16:29
|
|
|
 
|
|
#
?
Aug 27, 2015 16:30
|
|
|
Lol
|
|
#
?
Aug 27, 2015 16:31
|
|
|
get hungry
|
|
#
?
Aug 27, 2015 16:32
|
|
|
FaustianQ posted:I think our perspective has been reversed this entire time on the Nano/X relationship - Fury X is the dumping ground for lovely Fiji chips, considering the performance/watt gap is enormously in favor of the Nano despite being the same chip. It's why they need the CLC, it's why they came first. Someone needs to buy a Nano and slap a solid Liquid loop on it and see where they can take it, because if a CLC Nano outperforms an X then it all but confirms AMD marketing is run in a separate universe where it's normal to have your head shoved up your rear end. 
|
|
#
?
Aug 27, 2015 16:37
|
|
|
Anime Schoolgirl posted:But Fury X can't be the dumping ground for lovely Fiji chips because Fury already has a cut down chip so this is all sorts of confusing, how bad could the average Fiji bins possibly be? Well Intel does something similar with their CPU's. I believe the Core i7 is the 'standard' chip, with a Core i5 being a Core i7 that doesn't have functional hyperthreading, some defective cache or can't clock as well as the i7. Core i3's are Core i7's that have one or two defective cores. So in this case the Fury Nano is the Core i7, the Fury X is the i5 and the Fury is the i3.
|
|
#
?
Aug 27, 2015 16:46
|
|
|
Anime Schoolgirl posted:But Fury X can't be the dumping ground for lovely Fiji chips because Fury already has a cut down chip so this is all sorts of confusing, how bad could the average Fiji bins possibly be? Bins for good power/voltage performance are a different beast vs die harvested chips. GTX970 vs 980, Fury X vs Fury non-X, anything that's the same model of chip with functional bits disabled is a die harvest- It's totally possible that these still have excellent power performance profiles, it's just that there was a physical defect in one of the compute clusters. Intel is doing the power based bins on the latest broadwell Xeon E3s as well, where there's a 95w chip that has the exact same performance as a 65w chip. Beautiful Ninja posted:Well Intel does something similar with their CPU's. I believe the Core i7 is the 'standard' chip, with a Core i5 being a Core i7 that doesn't have functional hyperthreading, some defective cache or can't clock as well as the i7. Core i3's are Core i7's that have one or two defective cores. So in this case the Fury Nano is the Core i7, the Fury X is the i5 and the Fury is the i3. Not really true, dual core chips have entirely different dies. What you're thinking about happens more in the Xeon space, where there are 18 core chips and it's probably a lot more common to run in to a single core defect. Even there, though, the largest Xeons are completely distinct dies compared to the lower end 6/8/10/12 core versions. Gwaihir fucked around with this message at 16:55 on Aug 27, 2015 |
|
#
?
Aug 27, 2015 16:52
|
|
|
Beautiful Ninja posted:Well Intel does something similar with their CPU's. I believe the Core i7 is the 'standard' chip, with a Core i5 being a Core i7 that doesn't have functional hyperthreading, some defective cache or can't clock as well as the i7. Core i3's are Core i7's that have one or two defective cores. So in this case the Fury Nano is the Core i7, the Fury X is the i5 and the Fury is the i3. I thought that i3 / Pentium / Celeron were native dual-core chips, just like mobile i7 / i5 / i3s. Are desktop i3s really a full fat die?
|
|
#
?
Aug 27, 2015 16:54
|
|
|
Lol you guys, we virtually never talk about synthetics here except like literally this one time to diagnose a real problem
|
|
#
?
Aug 27, 2015 16:56
|
|
|
Beautiful Ninja posted:Well Intel does something similar with their CPU's. I believe the Core i7 is the 'standard' chip, with a Core i5 being a Core i7 that doesn't have functional hyperthreading, some defective cache or can't clock as well as the i7. Core i3's are Core i7's that have one or two defective cores. So in this case the Fury Nano is the Core i7, the Fury X is the i5 and the Fury is the i3.
|
|
#
?
Aug 27, 2015 16:57
|
|
|
The desktop pentium dual core/i3 are definately smaller physical chips compared to an i5/i7  So its not just lasering off dead cores
|
|
#
?
Aug 27, 2015 17:01
|
|
|
SlayVus posted:Try this Dropped to 18617, heh. It's fine, it can keep up with the games I'm playing lately, and I'm sure it'll get resolved. Thanks for the tip on the driver, though, I didn't know it was out there.
|
|
#
?
Aug 27, 2015 19:48
|
|
|
Apparently, Galax's office/warehouse is in Illinois. I just sold my 970 to a guy in Illinois. Just an interesting coincidence. FedEx says they'll deliver it Saturday. Just in time for my the days of from work. Come on Galax GTX 980 Ti HOF.
|
|
#
?
Aug 27, 2015 19:59
|
|
|
Heaven shows a meaningful difference between SLI and non-SLI, so it's not absolutely broken.
|
|
#
?
Aug 27, 2015 22:35
|
|
|
The windows 10 sli issue was a memory leak problem. The new 355.80 hotfix driver fixes it.
|
|
#
?
Aug 27, 2015 22:43
|
|
|
Don Lapre posted:The windows 10 sli issue was a memory leak problem. The new 355.80 hotfix driver fixes it. I think it's not the only issue, if I'm still getting 18K on firemark with dual 980 Tis and one of them is basically idle.
|
|
#
?
Aug 27, 2015 22:55
|
|
|
|
| # ? May 1, 2024 02:59 |
|
|
AVeryLargeRadish posted:Jesus Christ AMD, why? Just why? What the gently caress is the point of the Nano? Anyone interested in buying a loving $650 video card can also afford a big enough PSU to power a 275W card! Most of the ITX cases can fit such a PSU and a long, dual slot card! This thing is entirely pointless outside of a absolutely tiny demographic! gently caress, you can even get 600W SFX(-L?) PSUs nowadays. Silverstone just put one out. All you really need for a 300W card is about 450W, anyway, and those are readily available. The only catch there is heat - they're much smaller and 60-80mm is about all you can cram in there for a fan. So you have small fans spinning fast, which gets noisy. Yeah, short cards are an increasingly niche market because most cases are designed to take full-length double-slot cards. I have three different mITX cases and every single one of them could take a 980 Ti easily both in terms of power and space. I assume this will be reference only? Otherwise the AIBs would crank the clocks and TDP back up and slap a big cooler on it and boom custom Fury X boards. Paul MaudDib fucked around with this message at 23:47 on Aug 27, 2015 |
|
#
?
Aug 27, 2015 23:40
|
|
