|
So these are the good EVGAs? http://www.newegg.com/Product/Product.aspx?Item=N82E16814487090 http://www.newegg.com/Product/Product.aspx?Item=N82E16814487088 Hum 40$ premium for a backplate and higher factory overlclock.
|
#
?
Mar 12, 2015 00:45
|
|
|
|
| # ? May 29, 2024 07:55 |
|
|
sauer kraut posted:So these are the good EVGAs? I just looked this all up as well, and apparently that ACX 2.0+ cooler is an improvement but this MSI card will give you the same performance for $20-40 less if you just up a couple values in MSI afterburner. The overall cooling/power design is just flat better as far as I can tell. FWIW I'm basing that off this though, which is ~6 months old. The card settings don't seem at all different, just the available cooler for the EVGA cards. Mazz fucked around with this message at 01:18 on Mar 12, 2015 |
|
#
?
Mar 12, 2015 00:53
|
|
|
It should be noted though that EVGA's support/RMA is leagues ahead of ASUS and MSI.
|
|
#
?
Mar 12, 2015 01:05
|
|
|
GokieKS posted:It should be noted though that EVGA's support/RMA is leagues ahead of ASUS and MSI. Confirming this. I stuck to EVGA for my GTX 260 216 and GTX 760 (but went with MSI for the 970 this time around) and their RMA service is wonderful. I had a faulty 760 and they were awesome about taking it back for a replacement, no questions asked. Their step-up program is also pretty cool, though there is usually a long waiting list you have to deal with.
|
|
#
?
Mar 12, 2015 01:16
|
|
|
Subjunctive posted:I got the MSI, opened the Afterburner utility, moved the sliders all the way to the right, and was done. No problems so far (couple months). So like 1000+ mHz? 
|
|
#
?
Mar 12, 2015 01:21
|
|
|
veedubfreak posted:Adding voltage usually causes issues. Just use the power limit. To my knowledge, all current ACX 2.0 970s already have a 1.125v bump to voltage by default - he might actually need to *undervolt*. EDIT: Just "SSC" models and above, and it's a 1.215v boost. BIG HEADLINE fucked around with this message at 01:28 on Mar 12, 2015 |
|
#
?
Mar 12, 2015 01:21
|
|
|
One last question, I promise (no I don't): Are the 7-8gb versions of the 970/980 due to be out sometime in the next 30-60 days? If the price isn't extreme, is it worth waiting on those if I plan to keep it for a couple years and don't (currently) plan to SLI?
|
|
#
?
Mar 12, 2015 01:24
|
|
Mazz posted:One last question, I promise (no I don't): Are the 7-8gb versions of the 970/980 due to be out sometime in the next 30-60 days? If the price isn't extreme, is it worth waiting on those if I plan to keep it for a couple years and don't (currently) plan to SLI? 8gb is only going to be worth it at 4k resolutions in future games, and any current card is going to struggle a bit running at 4k on those games anyway. I'd just get a 970 if I was running at 1080p, what resolution is your current setup?
|
|
|
#
?
Mar 12, 2015 01:42
|
|
|
AVeryLargeRadish posted:8gb is only going to be worth it at 4k resolutions in future games, and any current card is going to struggle a bit running at 4k on those games anyway. I'd just get a 970 if I was running at 1080p, what resolution is your current setup? 1080p on a 27" Samsung with no real plans to upgrade any time soon as I randomly bought the monitor around New Years. Also wired to a 24" secondary monitor I run about half the time and a 50" TV if/when I'm watching something, but neither of those are relevant to the question really. I figured it wouldn't be an issue but I was thinking more about 1.5-2 years from now as games start eating up more VRAM. If it's only like 30 days off and $40-60 I can easily wait, my 480 is still functioning fine.
|
|
#
?
Mar 12, 2015 02:09
|
|
|
Mazz posted:One last question, I promise (no I don't): Are the 7-8gb versions of the 970/980 due to be out sometime in the next 30-60 days? If the price isn't extreme, is it worth waiting on those if I plan to keep it for a couple years and don't (currently) plan to SLI? nVidia (and even more so, the AIB makers) has little to no incentive at present to release something that'll just lead to the stagnation of people buying the 4GB 970s still out in channel. They probably won't even *show* them until Computex, and they probably won't be on sale until late August/early-to-mid September. Keep in mind nVidia had said they were going to release a driver that would help address and/or mitigate the memory 'issue' on the 970 by now. BIG HEADLINE fucked around with this message at 05:38 on Mar 12, 2015 |
|
#
?
Mar 12, 2015 05:33
|
|
|
How are they going to address the memory segmentation issue on the 970 if you stick 8gb of RAM onto it when the current hardware isn't designed to accommodate a full 4gb?
|
|
#
?
Mar 12, 2015 05:46
|
|
|
Ragingsheep posted:How are they going to address the memory segmentation issue on the 970 if you stick 8gb of RAM onto it when the current hardware isn't designed to accommodate a full 4gb? It's less that it can't do 4GB as that the last 8th is slow (and also slows down the first 7/8ths when you try to access it). So an 8GB 970 would be 7+1GB.
|
|
#
?
Mar 12, 2015 06:09
|
|
|
BIG HEADLINE posted:Keep in mind nVidia had said they were going to release a driver that would help address and/or mitigate the memory 'issue' on the 970 by now. That was debunked a long, long time ago. No driver update can fix the 970 situation but who cares, it's fast.
|
|
#
?
Mar 12, 2015 07:07
|
|
|
Also, I think the reason people are saying 7GB 970 is that they could theoretically have two pools of buffer that can pass through the bus at full speed (albeit one at a time), whereas an 8GB version would have two 3.5+0.5 pools.
BIG HEADLINE fucked around with this message at 07:44 on Mar 12, 2015 |
|
#
?
Mar 12, 2015 07:38
|
|
|
1gnoirents posted:whats the actual mhz? the offset isnt really indicative of much between cards and brands, but the final clock speed is far more absolute and what you should be comparing. Also you can simply max out your sandbox voltage and power target, you arent allowed to go over any damaging limit on these cards for a long time. Go into settings and select extended voltage settings as well its running at 1430mhz core on boost right now, is that good?
|
|
#
?
Mar 12, 2015 09:25
|
|
|
Ahdinko posted:its running at 1430mhz core on boost right now, is that good? Yeah that excellent for software OC. If it runs stable you don't need to do anything else.
|
|
#
?
Mar 12, 2015 12:00
|
|
|
Agreed that is fine. You won't be getting much more out of it due to plain chip stability although you can always push for a little more there. 1400-1500 is the norm
|
|
#
?
Mar 12, 2015 15:25
|
|
|
As with previous generations, doubling the memory on these cards is pointless. They are already limited by the actual ability to push pixels not the memory. With the announcement of the announcement in June from AMD and the Titan X you might keep an eye out for used 980s at this point. I picked up my pair for 450 each. Ended up splurging on a new motherboard yesterday. Open box Asus Maximus 7 Formula with 2 year warranty from Microcenter. Looks like whoever bought and returned it didn't even touch the thing. She's leak testing right now   I think this should scratch the itch until they finally get below 28nm.
|
|
#
?
Mar 12, 2015 15:39
|
|
|
veedubfreak posted:As with previous generations, doubling the memory on these cards is pointless. They are already limited by the actual ability to push pixels not the memory. Ah, static sensitive electronics on nylon carpet, what a good idea. Dude, at least put the bag in came in under it.
|
|
#
?
Mar 12, 2015 16:11
|
|
|
EoRaptor posted:Ah, static sensitive electronics on nylon carpet, what a good idea. Everything stayed on static bags. The motherboard tray is grounded. Ain't my first rodeo. My Gigabyte board died because I accidentally dropped the cpu and I think I bent a pin on the board while working on it a year or so ago. If any static was gonna kill something it would be the 2 furry "helpers" I have constantly roaming around me.
|
|
#
?
Mar 12, 2015 16:27
|
|
|
Haha I was about to say "at least there isn't a cat in the picture"
|
|
#
?
Mar 12, 2015 16:35
|
|
|
I thought you don't want to set stuff down on the outside of static bags because they divert static to the outside of the bag anyway.
|
|
#
?
Mar 12, 2015 17:18
|
|
|
beejay posted:I thought you don't want to set stuff down on the outside of static bags because they divert static to the outside of the bag anyway. The odds of actually hurting equipment in this day and age from static discharge is so small as to be almost negligible. I have worked on hundreds of computers over the years and never worn a static strap and have never damaged any equipment. It's not like I'm wearing flannel footie pjs while working on my stuff.
|
|
#
?
Mar 12, 2015 17:50
|
|
|
veedubfreak posted:The odds of actually hurting equipment in this day and age from static discharge is so small as to be almost negligible. This sounds really incorrect. I'm not an electronics guy, but the smaller semiconductor sizes should make equipment more vulnerable to static, no? High voltage flow through a small channel melts the poo poo out of it. I found this reddit post and others made by the same contributor incredibly enlightening as to the causes and symptoms of electrostatic discharge on computer components. To hear him tell it, thinking you have never damaged any equipment because you experienced no immediate failures is wrong: r00x posted:NO IT DOESN'T - ahhh gently caress, sorry, I'm just sick of people saying this. That's complete bullshit. It does NOT almost always cause immediate failure. You just think it does because you don't notice when it doesn't. There's also links to some sweet electron microscope images of damage.
|
|
#
?
Mar 12, 2015 19:25
|
|
|
Using ESD bands is so ingrained with me now, even if the risk was zero I'd probably find it hard not to use them when building machines  I've built a few machines, and some components failed within warranty, as you'd expect. (It'd be weird as hell if nothing ever failed) As I used ESD bands, I know that I observed the manufacturers guidelines when I built them, and I am blissfully blame free. We were taught that static (a looong time ago) can and will reduce lifespan, not that it just kills things stone dead immediately. (Efb!). GRINDCORE MEGGIDO fucked around with this message at 19:47 on Mar 12, 2015 |
|
#
?
Mar 12, 2015 19:29
|
|
|
Daviclond posted:This sounds really incorrect. I'm not an electronics guy, but the smaller semiconductor sizes should make equipment more vulnerable to static, no? High voltage flow through a small channel melts the poo poo out of it. I found this reddit post and others made by the same contributor incredibly enlightening as to the causes and symptoms of electrostatic discharge on computer components. To hear him tell it, thinking you have never damaged any equipment because you experienced no immediate failures is wrong: This guy seems to be talking about the dangers of ESD in the context of handling individual discrete electronic components not yet mounted on a PCB, which is a lot different than the dangers of ESD when handling a PCB that has proper ESD protection throughout. It's true that say a discrete semiconductor device like a MOSFET is very susceptible to damage from ESD but with good circuit design and proper PCB layout techniques everything can be protected to the point where the danger is pretty small. EDIT: Maybe I should test this out with an old GPU  
MaxxBot fucked around with this message at 19:59 on Mar 12, 2015 |
|
#
?
Mar 12, 2015 19:46
|
|
|
Daviclond posted:This sounds really incorrect. I'm not an electronics guy, but the smaller semiconductor sizes should make equipment more vulnerable to static, no? High voltage flow through a small channel melts the poo poo out of it. I found this reddit post and others made by the same contributor incredibly enlightening as to the causes and symptoms of electrostatic discharge on computer components. To hear him tell it, thinking you have never damaged any equipment because you experienced no immediate failures is wrong: Logically that's correct. Smaller processes should be more susceptible to damage than larger ones, thus one of the reasons why we still see what we think of as "ancient" processors used in high reliability applications. Practically my experience has been the same as veedubfreak's. I've had my hands in computers for over 20 years and haven't used any more anti-static precautions than "touch metal first" in that time without any problems that seem related. I had a CRT TV in the early '00s which would build up an immense charge on the screen, you could get 1/8-1/4 inch long sparks all day long by putting a finger near it. I took an old mobo/CPU/RAM I had around and decided to see what would happen if I placed it on the TV and zapped the poo poo out of it. I hit it all over the place with visible sparks and then hooked it up expecting to get either nothing or hilarious errors. Instead it booted up and ran normally. Memtest and Prime95 ran as normal, so I couldn't identify any damage. The last time I saw that board was a few years ago when my brother found it and used it to throw together a quick internet terminal for himself. It was still running seemingly fine. Slow of course, as it was a 200MHz Pentium MMX with 32MB RAM trying to run Firefox on Windows 98, but it worked. Now obviously minor damage could cause random bullshit I'd never be able to track down or identify a cause with, but it seems like overall static is one of those things that's way over-emphasized compared to the actual risk.
|
|
#
?
Mar 12, 2015 20:08
|
|
|
veedubfreak posted:The odds of actually hurting equipment in this day and age from static discharge is so small as to be almost negligible.
|
|
#
?
Mar 12, 2015 20:28
|
|
|
Just build your PCs naked and standing completely still.
|
|
#
?
Mar 12, 2015 21:29
|
|
|
Get yourself a humidifier, too. Edit: better still, assemble the PC underwater
|
|
#
?
Mar 12, 2015 21:32
|
|
|
Assemble it while submerged in mineral oil.
|
|
#
?
Mar 12, 2015 21:56
|
|
|
canyoneer posted:Just build your PCs naked and standing completely still. ph43r my 1337 ski11z? Largo would be proud.
|
|
#
?
Mar 12, 2015 22:02
|
|
|
How's the quality of the pasting between the 980HC and its OEM waterblock?
|
|
#
?
Mar 12, 2015 22:11
|
|
|
SwissArmyDruid posted:ph43r my 1337 ski11z? Largo would be proud. I... What... You should be ashamed of yourself (and so should I for getting it). How long am I going to have to wait for a 3gb 960? Or is it a "Spend that extra time waiting saving and get a 970" situation?
|
|
#
?
Mar 13, 2015 09:48
|
|
|
Gibbo posted:I... What... The latter. I'll be honest, I was amazed to find out earlier today that Megatokyo is, in fact, still running.
|
|
#
?
Mar 13, 2015 09:56
|
|
|
Not even a question, and if you wait too long the 970 might get superseded. Keep in mind other gaming events coming mid-year, and nVidia might feel it's milked Maxwell I enough by then. But an extra-RAM 960 is probably never going to be worthwhile.
|
|
#
?
Mar 13, 2015 09:59
|
|
|
Gibbo posted:I... What... There is rumors from China, but who knows http://linustechtips.com/main/topic/327161-next-generation-gpu-benchmarks-fiji-xt-390x-titan-x-cut-down-gm200-1080980-ti-960-ti965-rumor/ AMD Fiji quote:I want to know how much memory capacity, 8G, then pass, 16G surprise, 4G, then massage parlors to close it. quote:I regret to tell you, 4G Cut-down GM204 quote:There is no specific time? 970 and 980 is already on sale in the last six months of the profiteers and 965ti 960ti yellow on the warehouse is estimated to be moldy
|
|
#
?
Mar 13, 2015 10:02
|
|
|
Gibbo posted:How long am I going to have to wait for a 3gb 960? Assuming you want a 4GB one, it seems that Asus has confirmed that they will launch theirs this month. (But ^^^ it might not make sense to buy one.)
|
|
#
?
Mar 13, 2015 10:05
|
|
|
I think he means the rumored double-cut 980 that makes no sense for Nvidia to offer, not a weak 960 with more memory glued on.
|
|
#
?
Mar 13, 2015 10:07
|
|
|
|
| # ? May 29, 2024 07:55 |
|
|
Here's an interesting forum post on Mantle/Vulkan/DX12 by a somewhat well-known gamedev.quote:So why didn't we do this years ago? Well, there are a lot of politics involved (cough Longs Peak) and some hardware aspects but ultimately what it comes down to is the new models are hard to code for. Microsoft and ARB never wanted to subject us to manually compiling shaders against the correct render states, setting the whole thing invariant, configuring heaps and tables, etc. Segfaulting a GPU isn't a fun experience. You can't trap that in a (user space) debugger. So ... the subtext that a lot of people aren't calling out explicitly is that this round of new APIs has been done in cooperation with the big engines. The Mantle spec is effectively written by Johan Andersson at DICE, and the Khronos Vulkan spec basically pulls Aras P at Unity, Niklas S at Epic, and a couple guys at Valve into the fold.
|
|
#
?
Mar 13, 2015 10:15
|
|