|
Malloc Voidstar posted:How did he not realize he was using too much force? --edit: Also, at work I keep seeing people tightening the ever-loving poo poo on pieces that could not ever unscrew themselves. Apparently makes them feel better/safer. Anecdote: A coworker in another shift has tightened the same set of screws on a piece of machinery for like a hundred times with a torque wrench. So you'd figure he could loosely judge the amount of force needed. That one time that bumbling idiot accidentally switched the wrench to left-handed threads without knowing, he actually put a 5ft lever on the wrench, because it refused to click, and broke three screws until he gave it a second thought that something might be off. Three M20 screws. People are dumb. Combat Pretzel fucked around with this message at 17:44 on Aug 16, 2017 |
# ? Aug 16, 2017 17:28 |
|
|
# ? Jun 5, 2024 05:10 |
|
Computer building is pretty much like anything else: You can be strong or you can be smart. These guys right here chose to be strong today.
|
# ? Aug 16, 2017 19:42 |
|
AVeryLargeRadish posted:loving morons. loving goooons!! (It's yogscast lol)
|
# ? Aug 16, 2017 20:57 |
|
Oh good, then I won't have to go anywhere to let them know YOU UTTER PILLOCK, WHAT DID YOU THINK THAT 'CLUNK' COMING FROM THE WRENCH IN YOUR HANDS WAS ONCE THE SCREW WAS SNUGGED DOWN?!
|
# ? Aug 16, 2017 21:01 |
|
Does anyone here use Aida64? It seems pretty awesome and dicking around in the trial version seems swell. Although that $40 price tag for the home license I can certainly see off putting. What's the closest equivalent to Aida64 that isn't Prime95 and is free?
|
# ? Aug 16, 2017 21:20 |
|
Re: RIP Threadripper - LOL that's not inept, that's retarded.Combat Pretzel posted:M20 screws GRINDCORE MEGGIDO fucked around with this message at 21:26 on Aug 16, 2017 |
# ? Aug 16, 2017 21:21 |
|
Yea, I was surprised the wrench didn't give in (I mean, it's a sturdy 3ft long one, but still). We have a new wrench on our workbench now, though. It ratchets in the other direction now.
|
# ? Aug 16, 2017 21:54 |
|
Combat Pretzel posted:
I cringe when I see my cousin using anything involving a fastener that needs tightened. He will run a screw into the wall and let the driver strip/grind in the screw head for a good 10 seconds well after the screw has bottomed out. Ensuring there is no hope of anyone ever removing the screw in the future. And he works in construction, ugh...
|
# ? Aug 16, 2017 22:00 |
|
I can attest to people loving to over torque poo poo that really only needs to be a tad over hand tightened. Super common with fuckers racking servers and networking gear. I know its a couple hundred pounds but please, for the love of god, do not keep over tightening that poo poo...
|
# ? Aug 16, 2017 23:52 |
|
Does thermal compound ever go bad? The guy at Microcenter said this water cooling kit probably came with some so I didn't buy any but I have a ~15 year old tube of Arctic Silver around here... somewhere? Probably. If that stuff separates or something I'm not going to bother and just go back to Microcenter. Should have bought a case, anyway.
|
# ? Aug 17, 2017 00:27 |
stuff separating almost certainly means the emulsion settled-- its not an entropic phenomenon in the slightest. Just mix it back up! But regardless, any aio will come pre-pasted so you only need a tube if you gently caress it up or need to take it off again for whatever reason.
|
|
# ? Aug 17, 2017 00:49 |
|
Hah, it sure does have a square of grease on it. I didn't even think to look at it that hard I'm on the Ryzen bandwagon now E: this thing has a garish rainbow LED stripe that shines through a gap in the PCB but I have to go fishing for a motherboard speaker to see what the hell the beep code is when I get no video. Gamer culture is so loving trashy goddamn Munkeymon fucked around with this message at 02:20 on Aug 17, 2017 |
# ? Aug 17, 2017 01:03 |
|
Munkeymon posted:Does thermal compound ever go bad? The guy at Microcenter said this water cooling kit probably came with some so I didn't buy any but I have a ~15 year old tube of Arctic Silver around here... somewhere? Probably. If that stuff separates or something I'm not going to bother and just go back to Microcenter. Should have bought a case, anyway. Go bad? Ehh, not really. In AS3's case, because it's particulate suspended in an emulsified fluid, it *will* settle, and you'll get some super-dry compound as the first bit out of the tip. That's why AS3 recommends that you store it tip-down. So long as you've done that, just squeeze out the first dry bit, and then use as normal. (I have a foam block with holes punched into it to hold my syringes of goop, I just store them all that way, in case.)
|
# ? Aug 17, 2017 01:25 |
|
doomisland posted:I can attest to people loving to over torque poo poo that really only needs to be a tad over hand tightened. Super common with fuckers racking servers and networking gear. I know its a couple hundred pounds but please, for the love of god, do not keep over tightening that poo poo... What really got "proper torque settings" into my head was owning a Suzuki, made of compressed threaded cheese metal, and reading all the horror stories on bike forums of people trashing threads and needing helicoils, etc. I'm just pissy because I can't justify a TR at all and that guy just trashed one like a chimp.
|
# ? Aug 17, 2017 02:41 |
|
The chip is probably fine, socket might have taken some damage depending on if the screw head snapped off before the pins got overcompressed, but the motherboard is parts, now. Not that it's much of a consolation, what with being a $550 USD motherboard, BUT.
|
# ? Aug 17, 2017 02:44 |
|
Mofabio posted:Could I make an alternative suggestion for your CFD workload? Well it's worth thinking about. People have commented about memory bandwidth. For most CFD number crunching the CPU itself is usually the bottleneck as everything chokes up with the floating point calculations. When I was comparing the floating point throughput between Ryzen and the recent Xeons Ryzen is about 30% faster at the same clockspeed. There is a possibility that I might be wrong as some of the simulations I run might be broken down into such small workloads that the bottleneck may shift to cache or memory in whole or part. Although I have my doubt as the larger the model the complexity is exponential in terms of calculations required. I'll have to think about this a bit more and I'm not closing out any options.
|
# ? Aug 17, 2017 03:07 |
|
hmmmm
|
# ? Aug 17, 2017 03:25 |
|
Munkeymon posted:
what am i missing here?
|
# ? Aug 17, 2017 03:42 |
|
Scarecow posted:what am i missing here? My prime x370 is doing 8+8+8+8 and getting 16 as far as I can tell Or maybe it's only using half of it for some reason? At any rate, I need to look up some RAM overclocking guides because simply shoving more voltage in there isn't getting me the speed I bought
|
# ? Aug 17, 2017 03:52 |
|
Munkeymon posted:
Pablo Bluth fucked around with this message at 07:25 on Aug 17, 2017 |
# ? Aug 17, 2017 07:08 |
|
I really want to see what 3600mhz+ ram does for threadripper
|
# ? Aug 17, 2017 07:20 |
|
Taking a quad channel platfrom and only running 2 fast dimms instead of 4/8
|
# ? Aug 17, 2017 07:23 |
|
Risky Bisquick posted:Taking a quad channel platfrom and only running 2 fast dimms instead of 4/8 https://www.gskill.com/en/product/f4-4200c19q2-64gtzsw (god only knows how much that would cost)
|
# ? Aug 17, 2017 07:47 |
|
Did AMD fix their issue with RGB memory (SPD) btw?
|
# ? Aug 17, 2017 12:13 |
|
Nope, though it is more of a design flaw of the Trident RGB sticks than AMD's fault. Storing LED color settings in the SPD seems about as smart as having to re-flash your BIOS every time you change the CPU multiplier.
|
# ? Aug 17, 2017 12:25 |
|
Pablo Bluth posted:I had almost exactly the same issue last night when my memory arrived and I assembled the machine, except for 8/16 rather than 16/32. Turns out one of the dimms wasn't quite seated properly so it knew it was there but couldn't use it. Weird that it wouldn't say so explicitly on this otherwise overdone UI but, OK, I'll try reseating them when I get home tonight. I wish there were retention clips on both sides :\
|
# ? Aug 17, 2017 13:57 |
|
Scarecow posted:https://www.gskill.com/en/product/f4-4200c19q2-64gtzsw Godspeed to any moneybags that buy a 4 dimm DDR4000 kit
|
# ? Aug 17, 2017 15:05 |
|
Munkeymon posted:
Don't RAID 1 your RAM.
|
# ? Aug 17, 2017 18:30 |
|
AT corrected and retested their Game Mode results for ThreadRipper. http://www.anandtech.com/show/11726/retesting-amd-ryzen-threadrippers-game-mode-halving-cores-for-more-performance Honestly sorta looks like you are better off doing what AT did in the first place, and just turning off SMT. Actual Game Mode only seems to help with AMD cards, Nvidia cards got a performance decrease. And both brands were better in 16/16 mode than 8/16 Game Mode.
|
# ? Aug 17, 2017 19:23 |
|
They're still a bunch of chucklefucks. They mention DDR4-3200 RAM in the test setup, but apparently didn't test with it, or at least not list results. Reason? Being this:quote:While these comments make sense, ultimately very few users apply memory profiles (either XMP or other) as they require interaction with the BIOS, and most users will fall back on JEDEC supported speeds - this includes home users as well as industry who might want to shave off a cent or two from the cost or stay within the margins set by the manufacturer.
|
# ? Aug 17, 2017 21:15 |
|
That's especially ironic, since ian Cutress himself wrote a very clear, very opinionated guide about what XMP is and why people leave performance on the table by not enabling it.
|
# ? Aug 17, 2017 21:32 |
|
god what a loving stupid argument to try and make, "hey lets take a platform that has proven results that ram speed has a very real and direct impact on gaming benchmarks and run it at loving DDR4-2400 2400Mhz 3600Mhz CL16 from this video https://www.youtube.com/watch?v=RZS2XHcQdqA&t=215s
|
# ? Aug 17, 2017 23:24 |
|
Combat Pretzel posted:They're still a bunch of chucklefucks. They mention DDR4-3200 RAM in the test setup, but apparently didn't test with it, or at least not list results. Reason? Being this: So they didn't They didn't think to test at stock and with memory profiles changed as a comparison....... Dafuq
|
# ? Aug 18, 2017 00:47 |
|
I mean, to be fair, anything over 2666 is considered overclocking and warranty voiding for a lot of purposes. 2666 is the max JEDEC spec, and thats how the memory controller is officially listed. For doing an actual review of the supported product, it should probably be at 2666. Pragmatically, the type of people who read AT are the type of people who want to overclock (overclocking was how the website got popular ffs), but from a sperg perspective, that really is the product as it is offered. Its kinda like people that complain that car reviews are done on the tires from the factory.
|
# ? Aug 18, 2017 01:07 |
|
That's fine and good but have you considered por que no los dos?
|
# ? Aug 18, 2017 01:10 |
|
FaustianQ posted:That's fine and good but have you considered por que no los dos? cause it takes forever to run the full test suite and they dont get to keep review units forever, i imagine. article says: quote:Where possible, we will extend out testing to include faster memory modules either at the same time as the review or a later date. Not defending them completely though, they also posted the Ryzen review without any gaming results at all and poo poo. Theyve done lots of dumb stuff lately. I just dunno if running the memory at its officially supported speed is that dumb for a review of somethin.
|
# ? Aug 18, 2017 01:16 |
|
I can now confirm that a Ryzen 1700 is just a wee bit faster than an Athlon II X4.... I managed to get it running with the memory at 2800MHZ with just two mouse clicks and no understandings of what I was doing. The rated 3000 failed to boot so I'm going to have to RTFM.
|
# ? Aug 18, 2017 01:49 |
|
Pablo Bluth posted:I can now confirm that a Ryzen 1700 is just a wee bit faster than an Athlon II X4.... though 1T 2800 might be faster than 2T 2933 and if it's stable there after some stress runs i'd say go with it
|
# ? Aug 18, 2017 02:24 |
|
They ran the RAM at 2400 MHz and not 2666MHz if I understand the test setup page correctly.
|
# ? Aug 18, 2017 05:20 |
|
|
# ? Jun 5, 2024 05:10 |
|
Pablo Bluth posted:I had almost exactly the same issue last night when my memory arrived and I assembled the machine, except for 8/16 rather than 16/32. Turns out one of the dimms wasn't quite seated properly so it knew it was there but couldn't use it. Thanks for the advice! I reseated all of them which was annoying because the heat spreader is basically all sharp metal with one thumb-sized flat plastic bit to mock you, I guess? Anyway, memtest finished and I have 32 GB of probably-stable RAM running at 3200 w/the timings it's supposed to get. Motherboard is Asus PRIME X370-PRO and RAM kit (from the QVL) is https://www.newegg.com/Product/Product.aspx?Item=N82E16820232207 in case anyone else cares about shoving that much RAM in with boring old non-ripped dad threads.
|
# ? Aug 18, 2017 06:23 |