|
I slapped a noctua to my i7 920 in 2008 and only had to change fans since then. All fans eventually die or start making annoying noises.
|
#
?
Apr 27, 2019 02:42
|
|
|
|
| # ? May 16, 2024 21:44 |
|
|
Your system isn't in danger. It's very difficult to burn up a modern CPU. you will notice a performance drop and temperature hike. It sucks, but it's not like people need to pull these coolers out if everything is working fine.
|
|
#
?
Apr 27, 2019 02:42
|
|
|
Oh, if it's leaking, your system is in danger, alright. I'm just glad that materials science and engineering figured out how to make air coolers competitive again, because there's no way in goddamn hell I was clipping a 3 pound solid-copper fin stack onto socket A. These days though, I'm probably more liable to go with one of those expandable kits from EKWB or some poo poo. Cygni posted:Biostar confirmed in a round about way that the X570 boards are going to launch at Computex. Weirdly, AMD is also launching a "50th Anniversary Edition" 2700X next month (no clock speed bumps or anything, its just a 2700X). What, not even top .5% bins a la 8086K?
|
|
#
?
Apr 27, 2019 02:56
|
|
|
 maybe I missed some discussion on this but I thought those were filled with glycol or stuff that wouldn’t allow it to be filled up with organic goop.
|
|
#
?
Apr 27, 2019 03:10
|
|
|
Ethylene glycol concentrations above 25% don't kill bacteria, they just don't let bacteria grow and take over everything. Actually, I think the magic number is something like 22%, but the point is, if the concentration didn't hit 22% at minimum, it wouldn't stop goop from building up, if they did not add some other biocide to the mix.
|
|
#
?
Apr 27, 2019 03:20
|
|
|
With heatpipes you already get most of the benefits of water cooling without the downsides other than bulk directly on the CPU.
|
|
#
?
Apr 27, 2019 03:25
|
|
|
SwissArmyDruid posted:What, not even top .5% bins a la 8086K? Supposedly not :/ But you do get Lisa Su's signature burned into each one if you got the hots for her! 
|
|
#
?
Apr 27, 2019 03:25
|
|
|
Cygni posted:Supposedly not :/ Presumably you forgo a CPU cooler so you can show it off?
|
|
#
?
Apr 27, 2019 03:28
|
|
|
also either thats a fake or AMD's marketing department sucks so be skeptical. its from WCCF afterall
|
|
#
?
Apr 27, 2019 04:07
|
|
|
PC LOAD LETTER posted:It was really a LSU (Local Store Unit, aka scratch pad memory) and not like the L1 cache in a modern x86 chip (or one of that time period either). By "like L1 cache" I meant in terms of speed. Six cycles from memory to registers was in the order-of-magnitude of L1 latency of the PPU. PC LOAD LETTER posted:From what I recall of some of the comments from the guys over at B3D that actually had to make games on the thing that meant that pretty much all the memory management on the LSU's had to be done by the programmer by "hand" (compilers were supposed to help (since the SPU's supported branch hints) and did but never well enough to make up for the deficiencies of the LSU or lack of a proper branch predictor in the SPU's themselves), which was apparently quite difficult to do and was a major limiting factor in getting performance out of the SPU's. Particularly for the first few years the PS3 was out. Also the LSU's latency wasn't all that great (neither was the EIB's if you couldn't sustain a high level of concurrency with your work load which made things worse) and so any sort of cache misses or branch mispredicts were massively penalizing to performance (made even worse by the deeply pipelined and in order nature of the SPU's). I understand the system well, my PS3 code has shipped in a lot of games. The SPU had really short (6 cycle) pipelines, compared to the PPU (at 23). It was definitely less efficient than the PPU for general tasks, but the efficiency was mitigated by there almost always being more SPU capacity to spare, at least compared to the PPU which was always a bottleneck. I had a lot of fun fitting things onto the SPU that weren't what you would think would suit a co-processor. General purpose computing on GPU was still a few years out when the PS3 and Xbox 360 were conceived. The SPUs were better than DSPs for audio if only because console DSP hardware always seems to be a broken clusterfuck. The type of animation computation we did on the SPUs was a poor fit for GPU anyways, even once the consoles gained GPGPU capability that system never migrated.
|
|
#
?
Apr 27, 2019 05:09
|
|
|
priznat posted:
They just didn't get the mixture high enough it seems. However another problem is that when I took me cooler apart, it was actually pressurized. Nothing like what happened when Steve took theirs apart. Like, I took the fill port screw off the block and there was a spray of liquid on my face, the wall, and ceiling in my bathroom. I guess I was lucky to be wearing my glasses at the time. It probably would have really sprung a leak from the increased pressure and especially if it had still been in use and heated up over and over.
|
|
#
?
Apr 27, 2019 05:12
|
|
|
repiv posted:i had to dig up this old post because lmao "The Itanium approach...was supposed to be so terrific—until it turned out that the wished-for compilers were basically impossible to write." SlayVus posted:I switch to a ThermalRight Silver Arrow TR4 with an additional two Noctua NF-A12s two weeks ago. You know, I take a resolute stand that I don't give a poo poo about the appearance of my PC's guts. No RGB, no case windows, no painstaking "clean look" cable routing, the computer is just a box that sits on the floor and how the inside looks doesn't matter... But I gotta admit that cooler, with all the fat chrome heatpipe caps, looks real loving good. Like it's gonna make engine rev noises when you boot up. Klyith fucked around with this message at 06:04 on Apr 27, 2019 |
|
#
?
Apr 27, 2019 06:01
|
|
|
I find that not running wires in front of my fans or other components is both sexy and helpful That being said I also have a windowless case
|
|
#
?
Apr 27, 2019 12:48
|
|
|
Statutory Ape posted:That being said I also have a windowless case 
|
|
#
?
Apr 27, 2019 14:39
|
|
|
Statutory Ape posted:I find that not running wires in front of my fans or other components is both sexy and helpful 1: tangle of loose cables flopping through the middle of the case like it's 1999 in here 2: cables lying along the bottom of the case or bundled to the sides of the airflow, but still visible 3: fetishistic routing & wrapping of cables like you are the victorians and wires are ankles level 2 is useful, level 3 is aesthetic
|
|
#
?
Apr 27, 2019 20:30
|
|
|
Trying to route the wires cleanly usually starts out OK, until at some point, I start wanting to take my own life and give up on it. When I tossed the external fan controller in it, I just went with shortest path and have wires all over the place in the case. That said, I'd love to have a plain windowless case to begin with. Kinda like the NZXT H700 without window, but it's a poo poo case for airflow, too (removing the front panel dropped my water temps by more than 2.5°C, at idle).
|
|
#
?
Apr 27, 2019 20:45
|
|
|
I've found that a good case can really make the cable management easy, especially if you have a modular psu too. If you have a psu shroud or a basement it's the perfect place to stash cables, and then you just run everything else behind the motherboard tray
|
|
#
?
Apr 27, 2019 21:58
|
|
|
Interestingly, Gamers Nexus just posted a video about Enermax AIO gunk buildup: https://youtu.be/HC1kzO_gIp4 e: redundant post, my bad! The Illusive Man fucked around with this message at 03:00 on Apr 28, 2019 |
|
#
?
Apr 28, 2019 00:51
|
|
|
Space Racist posted:Interestingly, Gamers Nexus just posted a video about Enermax AIO gunk buildup: It's on the previous page which prompted the discussion.
|
|
#
?
Apr 28, 2019 01:03
|
|
|
I wonder if changing the coolant and adding a biocide (or just adding the biocide) right out of the package would help prevent buildup in the first place. I mean, you shouldn't have to, but right now there aren't that many other options, the Noctua isn't really intended for heavy overclocking (not sure why they didn't release a D15S variant for it too).
|
|
#
?
Apr 28, 2019 01:55
|
|
|
Stickman posted:Presumably you forgo a CPU cooler so you can show it off? How can I cook dinner if i use a cpu cooler? https://www.youtube.com/watch?v=wrmu_8LH88U
|
|
#
?
Apr 28, 2019 02:46
|
|
|
There's a bit in that GN video where he offhandedly says that maybe the rubber gasket is degrading. If that's the case (or if some metal parts are being eaten away) then more biocide won't help, or at best will delay symptoms. But it's hard to say because the big baby didn't want to touch anything gross, so he really didn't do much investigating. VostokProgram posted:I've found that a good case can really make the cable management easy, especially if you have a modular psu too. If you have a psu shroud or a basement it's the perfect place to stash cables, and then you just run everything else behind the motherboard tray My main PC case is still an Antec P182 so that probably has some bearing on my attitude. Over 10 years old and it still holds up for acoustics, but the cable management is primitive. But I have a spare PC in a modern corsair case, and I've build for friends in 2 fractal define variants. Even with those I don't route every cable through all the back panel holes and secret passages. The thing about putting absolutely everything in the back panel is that you then have to take the whole thing apart for even minor changes. And don't get me started on PSU shrouds, which are a pain in the rear end during the build, an even bigger PITA for swapping out a HD, and on many cases are a strict negative for thermals.
|
|
#
?
Apr 28, 2019 03:39
|
|
|
ehnus posted:By "like L1 cache" I meant in terms of speed. Six cycles from memory to registers was in the order-of-magnitude of L1 latency of the PPU. It is definitely lower than typical L2 caches of the time period but for a "L1" it was pretty high latency and in all other ways fell short of what a typical L1 cache did (ie. as far as I know it had no associativity, no store or miss queues, no tagging, no snooping, etc, all that had to handled or implemented or compensated or mitigated for somehow someway by the programmer) so I'd still say its not really reasonable to call it or compare it to a L1 cache. ehnus posted:I understand the system well, my PS3 code has shipped in a lot of games. By all accounts from everyone who had to ship something on it that was the most difficult platform to work with by far so if you had great success using it than that is drat impressive. I think your success might be causing you to look back at it with perhaps overly rosy tinted glasses though. Going by other developer comments who also used that hardware, and the fact that development and support for it disappeared as soon as was practical by Sony/IBM, it had far too many shortcomings to give its few rare real world advantages any real praise or positive consideration. ehnus posted:The SPUs were better than DSPs for audio if only because console DSP hardware always seems to be a broken clusterfuck.
|
|
#
?
Apr 28, 2019 04:00
|
|
|
I went from no side window case to my tempered glass sides phanteks and I love it 
|
|
#
?
Apr 28, 2019 04:05
|
|
|
I wonder if they are actually using deionized water or tap water.
|
|
#
?
Apr 28, 2019 04:13
|
|
|
Klyith posted:And don't get me started on PSU shrouds, which are a pain in the rear end during the build, an even bigger PITA for swapping out a HD, and on many cases are a strict negative for thermals. Don't even get me started on top facing I/O front ports.
|
|
#
?
Apr 28, 2019 04:14
|
|
|
Some potential zen2 leaks from a random chinese forum:
Overall, alright news I suppose. Really hoping for 4.7 all core on the high end x CPUs, but we'll see. Khorne fucked around with this message at 05:42 on Apr 28, 2019 |
|
#
?
Apr 28, 2019 05:36
|
|
|
Intel <$200 CPUs are already a tough sell as it is, and AMD getting another +7% IPC and 500MHz or so headroom while keeping current pricing tiers is going to turn that into an idiot check.
|
|
#
?
Apr 28, 2019 05:48
|
|
|
Space Racist posted:Stuff like this makes me never want to sway from air cooling. A NH-D15 may be huge but I can go years without having to even think about it. I'm way too paranoid about my PC to ever try water cooling. You don't have to worry about a NH-D15 springing a leak while you're away from home*. *Yeah, I run my PC 24/7 most of the time. Khorne posted:Some potential zen2 leaks from a random chinese forum: Interesting, although I'll believe 4.5Ghz on all cores when I see it first hand.
|
|
#
?
Apr 28, 2019 06:26
|
|
|
Paul MaudDib posted:I wonder if changing the coolant and adding a biocide (or just adding the biocide) right out of the package would help prevent buildup in the first place. I mean, you shouldn't have to, but right now there aren't that many other options, the Noctua isn't really intended for heavy overclocking (not sure why they didn't release a D15S variant for it too). It is pretty weird that they modified the 14s (and 12s!) and didn't bother with a TR4 D15s. What's your take on the best TR4 aircooling? The small amount of research I've done looks like it puts the Thermalright Silver Arrow on top, but there are probably some options out there that I'm not aware of.
|
|
#
?
Apr 28, 2019 06:38
|
|
|
Stickman posted:
|
|
#
?
Apr 28, 2019 07:14
|
|
|
Stickman posted:What's your take on the best TR4 aircooling? The small amount of research I've done looks like it puts the Thermalright Silver Arrow on top, but there are probably some options out there that I'm not aware of. I am underwhelmed by the U14S (and outright disappointed by every other cooler I saw before) but my understanding is that's the best cooler currently available. I hadn't seen results on the Silver Arrow; the money question is of course how it performs with the Noctua fans. Noctua's NF-A12x25 is so good that it completely distorts all results for coolers that aren't using it. Even their older fans beat the poo poo out of most of the stuff on the market. Right now the air coolers for TR4 are not that great; they are sufficient for maybe OC'ing a x950X or running a 2990WX at stock but they aren't excessive to the task at hand. If you really want to go hot you either get the AIO and deal with the gunk, or you get a waterblock and do a custom loop. If you're doing 2990WX I would prefer the custom loop for sure, you've got the money in the system and you should make the most of it. Like I said I don't get why they didn't do a D15S for the TR4 platform, if there's a platform that needs it that's the one. A D15S would probably have done better than a U14S as far as sales. Paul MaudDib fucked around with this message at 07:58 on Apr 28, 2019 |
|
#
?
Apr 28, 2019 07:20
|
|
|
Really can't wait for Zen2.. All those sexy PCIe Lanes without having to go Threadripper. Can finally get off my i5 2500K (which has served me pretty well considering its 8 years old now). We have all these bandwidth hungry standards (NVMe, Thunderbolt 3, USB3.2 on horizon) and still use PCIe standard from almost 10 years ago and less than 20 lanes. I want to be able to fit like eight NVMe drives in my computer some day.
|
|
#
?
Apr 28, 2019 07:47
|
|
|
Palladium posted:and AMD getting another +7% IPC and 500MHz or so headroom Appears that there won't be any sort of special AVX related clock speed reductions too which is also kinda cool.
|
|
#
?
Apr 28, 2019 09:30
|
|
|
PC LOAD LETTER posted:Supposed to be more like ~15%+ IPC over Zen+, peak all core clocks keep getting hinted at 4.5-4.7Ghz in the rumors so far (the latest clock speed leaks from WCCF/China are for what is rumored to be the 3600X) but single or dual core boosts to 5Ghz seem likely. Power usage and thermals at those speeds are supposed to be "pretty good" too, whatever that means. I can't wait for real info.
|
|
#
?
Apr 28, 2019 09:43
|
|
|
Khorne posted:Some of the OC guys said it can't go over 4.7 stable single core. This was a while ago so no idea how true it is. If it can get to 4.5Ghz all core without high temps/watts that is usually a good sign it should be able to do better. Usually near the peak clocks temp and power both start to skyrocket. Khorne posted:I can't wait for real info. edit: \/\/\/\/\/\/Me? No. The leakers are the ones making the claims which I'm essentially just repeating here, and lots of people who'd you expect to actually seem to know about the chip seem to be agreeing with them, or at least not saying they're BS like they normally do when a BS rumor pops up. So yeah they could be BS but at this point I kinda doubt it.\/\/\/\/\/\/ PC LOAD LETTER fucked around with this message at 10:45 on Apr 28, 2019 |
|
#
?
Apr 28, 2019 10:06
|
|
|
You got to 4.5 GHz all-core on a Zen2 sample?
|
|
#
?
Apr 28, 2019 10:13
|
|
|
Zen2 is going to have more available PCIe lanes than before? On the same socket? And they're coming directly from the CPU? How's that supposed to work?
|
|
#
?
Apr 28, 2019 11:49
|
|
|
Combat Pretzel posted:Zen2 is going to have more available PCIe lanes than before? On the same socket? And they're coming directly from the CPU? How's that supposed to work? The number of CPU PCIe lanes won't change. Hypothetically the number of lanes to the chipset from the CPU could change but probably won't. They're supposed to be using PCIe 4.0 so bandwidth to the chipset should be about double anyways. PCIe lanes from the B550 chipset (coming a couple of months or so later) probably won't change.
|
|
#
?
Apr 28, 2019 13:07
|
|
|
|
| # ? May 16, 2024 21:44 |
|
|
priznat posted:I went from no side window case to my tempered glass sides phanteks and I love it I went from no side windows on a Fractal Design R5 to tempered glass/RGB everything on a Lian Li PC-011 Dynamic  
|
|
#
?
Apr 28, 2019 16:02
|
|
