|
NewFatMike posted:I am extremely psyched to run a fat as hell Threadripper rig. I'm hoping there's minimal fuckery running a Windows VM for CAD and setting up the host machine as a rendering node. Imagine, compiling 128 packages all at once. 
|
#
?
Mar 6, 2019 21:56
|
|
|
|
| # ? May 18, 2024 10:19 |
|
|
NewFatMike posted:I am extremely psyched to run a fat as hell Threadripper rig. I'm hoping there's minimal fuckery running a Windows VM for CAD and setting up the host machine as a rendering node. and if you need render video vegas are amazing.
|
|
#
?
Mar 6, 2019 22:04
|
|
|
That's what I've heard from Level 1 Techs. Another fella in the firm handles animation/video rendering, but I'm already planning on taking all the product stills renders on my home machine. In my budget, AMD is kinda the only option GPU wise for VFIO pass through. I'm already planning on picking up the rumored ~$250 Navi GPU at 1080 performance (since I'll keep my 1080 for the host GPU), but if it's pricey for that performance, I might as well go whole hog on a Radeon VII for the guest GPU. I'm already afraid for my wallet for the Threadripper platform upgrade alone (and converting my R7 1700 into a NAS). I'm really hoping Navi spares me on the GPU front.
|
|
#
?
Mar 6, 2019 22:15
|
|
|
NewFatMike posted:I am extremely psyched to run a fat as hell Threadripper rig. I'm hoping there's minimal fuckery running a Windows VM for CAD and setting up the host machine as a rendering node. --edit: Then again, you're probably using Linux, anyway. I'm using VMs to isolate stupid poo poo from my host partition. Like Photoshop and Lightroom for instance. Combat Pretzel fucked around with this message at 23:02 on Mar 6, 2019 |
|
#
?
Mar 6, 2019 22:59
|
|
|
NewFatMike posted:the rumored ~$250 Navi GPU at 1080 performance This thought sounded so good until I remembered I’m on a Gsync monitor. gently caress. Here’s hoping that the 1080+ used market someday comes down parallel to the state of the 1060/1070 used market. I swear to god the pricing of Turing actually increased the used price of that poo poo.
|
|
#
?
Mar 6, 2019 23:27
|
|
|
MSRP/what I paid on my card was $699 when I bought it 13 months ago or so. 2 months after I bought it I could have sold it for near double. I had a gtx1060 that was selling for more used than what I paid for it new (purchased dec 2016) until recently Crypto mining is (was) a helluva drug E: prices have come down since Turing , but availability on 1080s has dwindled I think Worf fucked around with this message at 23:45 on Mar 6, 2019 |
|
#
?
Mar 6, 2019 23:42
|
|
|
Combat Pretzel posted:Graphics hardware acceleration is going to be an issue. If you're going to use Hyper-V, forget about it, because a) RemoteFX 3D has been deprecated and b) depending on the package, it may balk at it anyway (like Solidworks), and c) hardware passthrough only works with Quadro and poo poo. If you're going to use KVM, a cheap secondary graphics card to use with VFIO and Looking Glass should get you to places. Yeah, the second GPU+Looking Glass is the anticipated situation. Should be looking at the Beta 1 release in the next 12 weeks or so, which is well before I expect to see any Threadripper hardware or my ability to pay for it to catch up. Craptacular! posted:This thought sounded so good until I remembered I’m on a Gsync monitor. gently caress. Same on GSync, it looks like adaptive sync tech still holds even if you're mixing GPUs. I think the hardest part will be finding an EDID plug that'll let me pass through 3440*1440 at 90Hz.
|
|
#
?
Mar 7, 2019 00:04
|
|
|
pixaal posted:In my case, I'm the IT manager, and they are getting the same budget everyone else in design gets for their hardware so yes they will. Most places don't have someone in IT that actually understands the needs of a design department. So many people see a monitor is a monitor it doesn't matter if the color is accurate it just needs to be close enough. Which is pretty true in the business world, but if you are creating content you need the baseline to be good. At my workplace the user interface designers keep a few old lovely TN monitors around specifically in order to be able to test that the interfaces look okay on low end hardware.
|
|
#
?
Mar 7, 2019 02:32
|
|
|
Yay my new $650 threadripper arrived THE ACTUAL gently caress. E: At least the rest of the poo poo doesn't get here until friday, I have time to get them to resend me one overnight. Harik fucked around with this message at 03:59 on Mar 7, 2019 |
|
#
?
Mar 7, 2019 03:54
|
|
|
So I just picked up a 2700X to replace my current 1700X as I plan to build another PC and I figured I'd use this chance to do some marginal upgrades to mine. So far I'm really liking the XFR/PBO auto overclocking, it seems to pretty much make normal overclocking not necessary unless you're demanding very high all-core locked clock speeds. For gaming purposes, it's much more useful for me to have a couple of threads boost up while the others stick at slightly lower clocks. But I do have a question about the Performance Enhancer stuff on my Asus Crosshair VII Wi-Fi. My understanding is that levels 1-2 are AMD approved settings and shouldn't be an issue, level 3 and level 4 are Asus' own stuff. I do have a 280mm AIO cooler and in my testing it looks like it can handle Level 3 just fine temps wise, I was just wondering if I'm being paranoid about it potentially being dangerous to the CPU to let the mobo do what it wants with the voltages to keep things clocked as high as possible? Level 4 seems to be reserved for people doing exotic cooling and I think I'll steer clear of that for now. Other than that, I'm actually pretty happy with the upgrade so far. My 1700X wasn't a great overclocker, only did 3.8 before I was pushing past the recommended voltage limits of overclocking Ryzen. The increase in performance going from that to a 2700X doing 4.1-4.2ghz boosting up is actually noticeable for me, it's cleared up a lot of CPU related stuttering I was having before, I've moved on to an ultrawide monitor recently and didn't take into account that my CPU now needs to draw an extra 25% or so objects on screen, my 1700X was just not keeping things as smooth as I liked once I made the switch. Really looking forward to the rumored 3700X if that ends up being a thing, something that can boost to 5ghz on a couple cores would be super sweet for gaming.
|
|
#
?
Mar 7, 2019 06:12
|
|
|
Cygni posted:Why would Apple move to AMD when they can just threaten to do it every couple of years (and even build and "leak" obvious AMD powered examples on public benchmakring sites), and get Intel to sell them CPUs cater made to their specifications for a loss? Well, the new speculative execution exploit that appears to ONLY affect Intel processors, and needs mitigation at the silicon level, that was recently disclosed might be a start.
|
|
#
?
Mar 7, 2019 06:28
|
|
|
Harik posted:Yay my new $650 threadripper arrived yikes! (it's probably fine? threadrippers are LGA so no pins to be bent. I feel like rattling around in there a bit is something it could take without damage, if the box isn't totally smashed. not that it's worth finding out on a $650 item.)
|
|
#
?
Mar 7, 2019 06:44
|
|
|
Its that orange plastic bracket that I'd worry about. You need it intact so that you can put the CPU into the socket without bending the LGA pins.
|
|
#
?
Mar 7, 2019 06:53
|
|
|
Harik posted:Yay my new $650 threadripper arrived ThreadRIPper
|
|
#
?
Mar 7, 2019 07:22
|
|
|
wargames posted:probably not, apple is probably going to go with an arm solution and move away from x86-64 so they can have a "one os" walled garden. There's a limit to how walled the garden can get. If Apple doesn't have a development platform then there's no new or updated software going into the App Store. And Apple tried to make OS X into a server platform, if you need a hint as to how far Apple's willing to go to keep everything in-house, so I'd be rather surprised if they use someone else's general-purpose platform for software development. But this doesn't rule out a general-purpose ARM-based macOS, which probably wouldn't be so bad, or aggressive development controls (along the lines of background-check+upfront+subscription console-style dev kits being at some future time the only GP Apple platforms left), which definitely would. dont be mean to me fucked around with this message at 11:41 on Mar 7, 2019 |
|
#
?
Mar 7, 2019 11:38
|
|
|
Harik posted:Yay my new $650 threadripper arrived I thought the packaging on mine (altho a 2950X) was obnoxious as gently caress. How could that happen to a new one? Hell, I don't even have anything of it anymore, except that orange clamshell thingy where the actual CPU was in. Gonna be a fun one some people will probably whine about when I want to ditch it on Ebay or whatever.
|
|
#
?
Mar 7, 2019 12:52
|
|
|
Combat Pretzel posted:You bought that new or used? That's new. I didn't even break the seal, I just sent it back. As for how it could happen, well:  I'm the half-assed strip of bubble wrap taking a journey of discovery through the carnival rollercoaster of domestic shipping.
|
|
#
?
Mar 7, 2019 15:27
|
|
|
I bet its fine, pop it in coward
|
|
#
?
Mar 7, 2019 15:29
|
|
|
IIRC there's SMD parts on the bottom. Maybe some went missing with all the tossing about?
|
|
#
?
Mar 7, 2019 19:35
|
|
|
 I remember watching unboxings of this and thinking that this was so goddamn overkill. The chip is mounted in a plastic pod, which is locked into the foam outer casing, and it still managed to pop apart and bounce around to death? edit: At any rate, I forgot why I checked in the thread in the first place. I found some interesting things in the investor relations slide deck.   SwissArmyDruid fucked around with this message at 20:14 on Mar 7, 2019 |
|
#
?
Mar 7, 2019 19:47
|
|
|
Cool, with the new Intel vulnerabilities it's probably a safe bet that Zen2 will be in high demand, leading to inflated launch prices and low supply. And here I was, thinking I could get a good deal to finally replace my 2500k...
|
|
#
?
Mar 7, 2019 22:48
|
|
|
Does up to 2x performance per socket mean that the fastest CPU is 100% faster than that of the previous generation or something else I’m misunderstanding? Edit: I’m an idiot and thought that infographic was about zen 3. Fenom fucked around with this message at 22:53 on Mar 7, 2019 |
|
#
?
Mar 7, 2019 22:51
|
|
|
Fenom posted:Does up to 2x performance per socket mean that the fastest CPU is 100% faster than that of the previous generation or something else I’m misunderstanding? Short answer it doesn't mean anything, it's just some marketing speak. Long answer, it's likely a combination of more cores per CPU and process improvements leading to a doubling of performance in some very specific metric.
|
|
#
?
Mar 7, 2019 22:54
|
|
|
They are doubling core counts per socket across the range (in all likelihood), so thats probably what they are talking about.
|
|
#
?
Mar 7, 2019 22:59
|
|
|
Harik posted:That's new. I didn't even break the seal, I just sent it back. As for how it could happen, well: Shippers must have made a point to field goal kick the living poo poo outta that particular package. Best of luck to you on the next shipment.
|
|
#
?
Mar 7, 2019 23:06
|
|
|
Beautiful Ninja posted:So I just picked up a 2700X to replace my current 1700X as I plan to build another PC and I figured I'd use this chance to do some marginal upgrades to mine. You don't need anything exotic for PE4. I'm running a 360mm AIO and it handles PE4 fine. I run a -0.09375 voltage offset with mine and it keeps the voltage around 1.4 or lower under load. Some people can pull off a -0.1 offset but mine won't. One thing I've seen somewhat commonly is PE3 with BCLK adjustments and a positive offset. I haven't played around with that yet but will at some point in the future.
|
|
#
?
Mar 8, 2019 01:10
|
|
|
mcbexx posted:Cool, with the new Intel vulnerabilities it's probably a safe bet that Zen2 will be in high demand, leading to inflated launch prices and low supply. We’ve been through so many of these, including Meltdown and a bogus AMD one, and it’s never made a significant impact on sales. The average person is willing to let Intel, Microsoft, and the assorted big data engineers figure it out for them, and isn’t in a hurry to replace hardware.
|
|
#
?
Mar 8, 2019 01:45
|
|
|
Craptacular! posted:We’ve been through so many of these, including Meltdown and a bogus AMD one, and it’s never made a significant impact on sales. If some new virus popped up and stole a few million credit card numbers using these methods then you'd see it make a impact on sales. At least on the consumer side of things. On the server side of things I think it matters more and there does seem to be at least a little more interest in newer AMD hardware because it seems to be effected little or not at all by any of these new exploits which just keep popping up over and over now. Everyone seems to be waiting for the next shoe to drop so to speak.
|
|
#
?
Mar 8, 2019 02:05
|
|
|
I think there are two major related reasons most of these attacks seem to be affecting Intel much more than AMD. Firstly, this is a new class of attacks discovered only a year ago. There is still a lot of ground security researchers have yet to cover. Secondly, between 2012 and 2017 AMD was pretty much completely out of the CPU game, especially in the enterprise market. Consequently, there are a lot more Intel CPUs in the wild to attack and a lot more security researchers who have access to Intel CPUs to experiment with, so it's natural that the early research efforts will focus on Intel CPUs. The SPOILER attack authors noted they tested 10 Intel CPUs but only 1 AMD CPU (and a Bulldozer-based APU at that). Because AMD's Zen architecture and Intel's Core architecture are only loosely related (in that they consume the same instruction set) it probably shouldn't be a surprise that attacks on one mostly don't affect the other. It's likely that there are major attacks lurking within AMD CPUs, there's just not enough attention being focused on AMD CPUs to discover them. That's not to say that AMD CPUs can't genuinely be more secure than Intel CPUs, but at this point we don't have enough information to say for sure.
|
|
#
?
Mar 8, 2019 02:32
|
|
|
Balliver Shagnasty posted:Shippers must have made a point to field goal kick the living poo poo outta that particular package. Best of luck to you on the next shipment. I love your av. Did you make the horsegame it came from? I vaguely remember that thread. Boringly, round 2 arrived intact, but they did manage to kick the Define S2 box hard enough to rip it and leave a dent in the front panel. 2 for 2, can't wait to see the state the rest of my stuff arrives in tomorrow.
|
|
#
?
Mar 8, 2019 02:33
|
|
|
fknlo posted:You don't need anything exotic for PE4. I'm running a 360mm AIO and it handles PE4 fine. I run a -0.09375 voltage offset with mine and it keeps the voltage around 1.4 or lower under load. Some people can pull off a -0.1 offset but mine won't. One thing I've seen somewhat commonly is PE3 with BCLK adjustments and a positive offset. I haven't played around with that yet but will at some point in the future. Thanks for that, I'll start working on finding a good offset for my CPU. While looking at voltages, I noticed with PE level 3 using Auto, during heavy CPU stuff like Cinebench my voltages were only around 1.32 or so while the CPU stuck at 4.1ghz. But I did notice under more lightly threaded work, like some games, my voltages went up to 1.5v when 1-2 cores would sit at 4.3 while others are at lower clocks. My suspicion is that this is actually intended when PBO is clocking a couple cores high as Ryzen will need those volts to get the clocks that high, but I just wanted to double check. Next on the agenda is getting my RAM working at a good speed. Ended up buying some DDR4 4000 because it was relatively cheap for Samsung B-Die. Obviously it failed miserably at loading its XMP profiles at both 3600 and 4000, right now I'm running it at 3200 CL14 because it doesn't require me to put more than the XMP standard 1.35v onto the memory. My understanding is that up to 1.5v is fine here and probably needed to push into the high DRAM frequency range, looking to see if I can push it up to 3600 since Ryzen DRAM calculator says I should be able to do that at relatively low timings.
|
|
#
?
Mar 8, 2019 04:34
|
|
|
Mr.Radar posted:Firstly, this is a new class of attacks discovered only a year ago. There is still a lot of ground security researchers have yet to cover. Secondly, between 2012 and 2017 AMD was pretty much completely out of the CPU game, especially in the enterprise market. Consequently, there are a lot more Intel CPUs in the wild to attack and a lot more security researchers who have access to Intel CPUs to experiment with They target computing features (ie. speculative execution) not specific cores or ISA's which is why even stuff like ARM as well as VIA's x86 CPU's and Apple's various custom processors are effected by Meltdown and Specter and AMD's Zen is only somewhat vulnerable to Specter but immune to Meltdown and SPOILER. So implementation specifics do indeed matter but its unreasonable to claim that because AMD's marketshare with Bulldozer was low researchers just never bothered to test Zen or Zen+ or even BD itself for similar vulnerabilities. Researchers have indeed been testing Zen/Zen+ and Bulldozer (they often state this publicly) for these vulnerabilities and BD (and even older stuff like Phenom/II and A64) has been found to be susceptible to them while Zen/Zen+ hasn't. Its also worth pointing out that BD and older AMD chips, while selling poorly, weren't hard or expensive to get for testing. Used or new. If anything the poor sales of BD forced AMD to compete on price by drastically slashing prices on most parts so that technically would've made it easier for researchers to get the parts new if the wanted. Used they're even cheaper. And Bulldozer mobo's and CPU's, or entire systems, could be easily bought at many retailers despite their poor marketshare for years. Or even now really. B&M or online, though now mostly online. PC LOAD LETTER fucked around with this message at 04:49 on Mar 8, 2019 |
|
#
?
Mar 8, 2019 04:46
|
|
|
Harik posted:I love your av. Did you make the horsegame it came from? I vaguely remember that thread. See, this is why I don't feel bad about driving 3 hours to the nearest Micro Center to pick my parts in person. Any fuckups that happen on the ride back are mine and mine alone. Sadly, I didn't make the horse game. In fact, I practically forgot how and where I even got that AV from.
|
|
#
?
Mar 8, 2019 04:58
|
|
|
Beautiful Ninja posted:Thanks for that, I'll start working on finding a good offset for my CPU. While looking at voltages, I noticed with PE level 3 using Auto, during heavy CPU stuff like Cinebench my voltages were only around 1.32 or so while the CPU stuck at 4.1ghz. But I did notice under more lightly threaded work, like some games, my voltages went up to 1.5v when 1-2 cores would sit at 4.3 while others are at lower clocks. My suspicion is that this is actually intended when PBO is clocking a couple cores high as Ryzen will need those volts to get the clocks that high, but I just wanted to double check. I think a lot of the issues people have with PE4 is that they aren't running a negative offset on it or are running air cooling. I updated to the latest BIOS yesterday and had to reset everything and ran an hour of prime95 and my temps maxed out at 70 degrees. If I remember correctly from when I first did PE4 without an offset I was hitting 80+. The voltages are pretty high on it and my goal was to get it below the 1.425 that I've seen recommended as the max regular voltage. I was generally below 1.4 on the last BIOS and seem to be right around/just above it on the latest one. PE3 obviously runs much cooler but also obviously doesn't hit the same all core speeds.
|
|
#
?
Mar 8, 2019 05:12
|
|
|
Beautiful Ninja posted:Thanks for that, I'll start working on finding a good offset for my CPU. While looking at voltages, I noticed with PE level 3 using Auto, during heavy CPU stuff like Cinebench my voltages were only around 1.32 or so while the CPU stuck at 4.1ghz. But I did notice under more lightly threaded work, like some games, my voltages went up to 1.5v when 1-2 cores would sit at 4.3 while others are at lower clocks. My suspicion is that this is actually intended when PBO is clocking a couple cores high as Ryzen will need those volts to get the clocks that high, but I just wanted to double check. I'm still wondering if I should stick with my all-core OC or go back to XFR/PBO. One of the oddities of my current OC is that clock speeds don't drop below 3.9Ghz, even with Cool n' Quiet and C-states enabled. This might answer my own question, but an all-core boost seems like a waste if the vast majority of your programs and games use only one or two cores. As for RAM, I've heard that getting speeds past 3400 on a Ryzen chip is a gamble, hence why I didn't bother getting RAM any faster than 3200. Plus my sticks are SKHynix A die, but I had absolutely no trouble getting up to 3200 with XMP. What you're doing with the voltage and timings seems like a good idea imho.
|
|
#
?
Mar 8, 2019 05:13
|
|
|
Balliver Shagnasty posted:As for RAM, I've heard that getting speeds past 3400 on a Ryzen chip is a gamble, hence why I didn't bother getting RAM any faster than 3200. Plus my sticks are SKHynix A die, but I had absolutely no trouble getting up to 3200 with XMP. What you're doing with the voltage and timings seems like a good idea imho. It's not just a gamble, it's that the Zen cores are both latency and bandwidth sensitive so CL12 3200 is actually optimum IIRC, with CL14 3600 being close. DDR4 4000 to 4400 tend to have much looser timings and thus even if achievable can have a net negative. The rare circumstances in which you can achieve latency/clock parity you get similar results as CL12 3200 and CL14 3600, you basically need an unachievable CL14 4000 to see performance improvement. I'm really interested in how Zen2 handles the IF/Memory. Could a 1:1 divider mean Zen2 cares less about clockspeed and just wants the lowest latency possible?
|
|
#
?
Mar 8, 2019 17:33
|
|
|
mcbexx posted:Cool, with the new Intel vulnerabilities it's probably a safe bet that Zen2 will be in high demand, leading to inflated launch prices and low supply. if you want bargains, 2xxx series ryzens are already starting to have firesales Mr.Radar posted:I think there are two major related reasons most of these attacks seem to be affecting Intel much more than AMD. Firstly, this is a new class of attacks discovered only a year ago. There is still a lot of ground security researchers have yet to cover. Secondly, between 2012 and 2017 AMD was pretty much completely out of the CPU game, especially in the enterprise market. Consequently, there are a lot more Intel CPUs in the wild to attack and a lot more security researchers who have access to Intel CPUs to experiment with, so it's natural that the early research efforts will focus on Intel CPUs. The SPOILER attack authors noted they tested 10 Intel CPUs but only 1 AMD CPU (and a Bulldozer-based APU at that). All 100% true -- though I don't think that security researchers have a hard time getting their hands on AMD cpus. I think they just don't bother extensively testing AMD cpus for attacks crafted for Intel, for the exact reason you point out: they know it won't work with a completely different microarchitecture. However, I think there's at least one reason to suspect that AMD is less vulnerable to these types of exploits: it's an area where their stuff is much less sophisticated than Intel's. The quality of Intel's prefetch, speculative exec, branch predictors, etc was a major advantage for real-world IPC and performance. Until now. I don't think it's totally a coincidence that Intel is seeing more vulnerabilities when their architecture is more complex.
|
|
#
?
Mar 8, 2019 19:24
|
|
|
Klyith posted:if you want bargains, 2xxx series ryzens are already starting to have firesales For good reason, the 3rd gen
|
|
#
?
Mar 8, 2019 19:55
|
|
|
Risky Bisquick posted:For good reason, the 3rd gen So desperate for concrete info on it.
|
|
#
?
Mar 8, 2019 20:01
|
|
|
|
| # ? May 18, 2024 10:19 |
|
|
Klyith posted:if you want bargains, 2xxx series ryzens are already starting to have firesales Eh, I have waited so long, I wanna see what the 570x chipset brings other than PCIe 4.0, do one big upgrade and be done with it for the next 5+ years (got the 2500k in... 2012?). Stopgapping with a 2700x for 4-6 months doesn't make sense to me. Just don't delay anymore pls. Also, don't gently caress up the launch.
|
|
#
?
Mar 8, 2019 23:09
|
|