|
Purgatory Glory posted:Wow, I clicked purchase expecting to be taken to a link calling me a gullible retard. Instead it really asked me for $159.00 for 3 aquarium rocks in a bag. Machina Dynamica started as a joke, but people started sending them money. It's much more obviously a joke now with products like this -- but check out some of their earliest pages in the Wayback Machine. To provide context, isolation platforms for speaker cable are actually a thing audiophiles buy. The theory being that your cables being on the ground increases microphonics from vibrations and increases the capacitive effect of the insulation. You can find tons of testimonials about how much of a difference these make at the margins! I am actually having a ton of trouble trying to find a place that sells these, but you can see a picture of what I'm talking about here: Audiophiles go through fads, and around the time Machina Dynamica popped up these were super popular -- and on the theory that more expensive is better -- sites respected by the community were absolutely selling versions of these for $100+ each. EDIT: Linked the wrong product, having trouble finding the exact one I am thinking of -- which were platforms for just the cables and not your entire rig. Isolation platforms are definitely still a thing people buy -- mainly for vinyl. You can easily spend thousands on these, with some fancy ones based on extreemly strong magnets and ferrofluids. Meanwhile -- research scientists that need to isolate highly sensitive equipment from vibrations use platforms floating in pits of sand. A $20 solution. Chuu fucked around with this message at 20:27 on Jul 3, 2015 |
# ? Jul 3, 2015 19:33 |
|
|
# ? Jun 5, 2024 05:15 |
|
Chuu posted:Yeah, I was exaggerating a bit, and I was mainly meaning the analogue stuff. I spent some time building amps and involved with the DIY audiophile croud, and some of the circuits people swear by; especially all the Class A stuff; is from the 80's or earlier. Some of it much earlier. Youll pry my class a amps from my cold dead hands
|
# ? Jul 3, 2015 19:46 |
|
Panty Saluter posted:Youll pry my class a amps from my cold dead hands class a amps are dumb as hell, it's literally burning power class D amps are sw8 and could be everywhere except for AUDIOPHILES
|
# ? Jul 3, 2015 19:54 |
|
I, too, am completely oblivious to switching noise
|
# ? Jul 3, 2015 19:59 |
|
If I get a magnetic levitating monitor stand will I have less screen tearing issues?
|
# ? Jul 3, 2015 20:08 |
|
No, also there's a reason there's vesa arms in all varieties.
|
# ? Jul 3, 2015 20:13 |
|
If ya'lls signal is dirty, try washing your cables.
|
# ? Jul 3, 2015 20:39 |
|
https://www.pugetsystems.com/labs/articles/Thermal-Paste-Application-Techniques-170/ I think thermal paste voodoo is the closest thing we have to random-rear end pseudoscience. I love how the author tries 20 different patterns and insists on a best one, even though his own chart shows that they're pretty much all the same.
|
# ? Jul 3, 2015 21:43 |
|
INTJ Mastermind posted:https://www.pugetsystems.com/labs/articles/Thermal-Paste-Application-Techniques-170/ A 2C difference under load simply from using a different pattern of the same paste seems reasonable enough. I mean, people chase similar gains when arguing about various coolers. At least it's something that can be actually measured empirically.
|
# ? Jul 3, 2015 21:51 |
I've always done smooth spread myself probably based off of advice I got in 1998. So it's good to see that it's still effective. I did have an Athlon XP that scorched through the thermal pad and ran bare metal on metal for what could have been years as far as I could know, on an ECS board, no less. The takeaway I got from that is the spread probably isn't so important, so long as the HSF is on there tight enough. And to never trust the pad on a stock fan if it's got one. It's been all Ceramique for me from that point on for both GPUs and CPUs.
|
|
# ? Jul 3, 2015 22:13 |
|
DrDork posted:At least it's something that can be actually measured empirically. And that's the crucial difference between this and audiophiles.
|
# ? Jul 3, 2015 22:56 |
|
I like how MSI do theirs. This was from their official channel before they took it down for some reason? https://www.youtube.com/watch?v=kSCVahyh3Dk
|
# ? Jul 3, 2015 22:56 |
|
INTJ Mastermind posted:https://www.pugetsystems.com/labs/articles/Thermal-Paste-Application-Techniques-170/ Well basically you're trying to prevent air bubbles so good luck with a less effective pattern can outperform a usually better one.
|
# ? Jul 3, 2015 23:05 |
|
Ak Gara posted:I like how MSI do theirs. This was from their official channel before they took it down for some reason? I too use toothpaste as my TIM of choice.
|
# ? Jul 3, 2015 23:08 |
|
DrDork posted:A 2C difference under load simply from using a different pattern of the same paste seems reasonable enough. I mean, people chase similar gains when arguing about various coolers. At least it's something that can be actually measured empirically. If he was logging environmental temperatures too and if his conclusions made any loving sense I might give a poo poo what he did. A loving happy face outperforming the rough spread, three thin lines, spiral pattern etc leaves me quite convinced that his results are more variance in heatsink application technique than anything to do with the pattern. Ak Gara posted:I like how MSI do theirs. This was from their official channel before they took it down for some reason? Did MSI buy out Arctic Silver or something? That poor CPU nearly drowned.
|
# ? Jul 3, 2015 23:20 |
|
lmao it is a fuckton, I thought I drowned my poo poo.
|
# ? Jul 3, 2015 23:35 |
|
K8.0 posted:If he was logging environmental temperatures too and if his conclusions made any loving sense I might give a poo poo what he did. A loving happy face outperforming the rough spread, three thin lines, spiral pattern etc leaves me quite convinced that his results are more variance in heatsink application technique than anything to do with the pattern. Three thin lines, spiral and rough spread all seem like a great way to get air bubbles in the TIM.
|
# ? Jul 3, 2015 23:47 |
|
Ak Gara posted:I like how MSI do theirs. This was from their official channel before they took it down for some reason?
|
# ? Jul 3, 2015 23:47 |
|
Just to prove that it's really difficult to gently caress up thermal paste...when my sophomore roommate was building his first custom computer, he spread out Radio Shack-branded thermal paste onto the bottom of his HSF assembly *with his unwashed finger*, adding none to the CPU. He later had to redo it and did it properly...and he got better temps with the finger-spread Radio Shack paste than he did doing it 'properly' the second time with Arctic Silver.
|
# ? Jul 4, 2015 00:04 |
|
Also, paste application is CPU dependent, according to Arctic Silver, anyway.
|
# ? Jul 4, 2015 00:59 |
|
It wasn't that long ago that image quality mattered, like the generation before one of 'em (brain's telling me AMD) ripped off the other one's AF algorithm more or less entirely and it's been pretty neck and neck since then. Both of them have solid AA tech, no real difference there anymore apart from the details. Nobody seems to care about AA anymore, though, it must all be good enough now that resolutions are getting higher for enthusiasts at least and perf is better carved out of other stuff. I'm really out of the loop on all this crap now, I still like my 780Ti and I haven't even played the witcher 3 since it came out and my preorder turned into a game. I have however beat Crimsonland and also Red Nation on my laptop, which has a GTX 540m in it and overheats regularly. The point is, buy AMD
|
# ? Jul 4, 2015 01:27 |
|
One Eye Open posted:Also, paste application is CPU dependent, according to Arctic Silver, anyway. I'm sure the difference between a horizontal and vertical pattern are very thermally significant
|
# ? Jul 4, 2015 01:38 |
|
Intel could fix this simply by putting a thin plastic cover over their heat spreader-covered CPUs that demarcate exactly where the core rests beneath it. Covering the *entire* spreader plate isn't necessary, since the bulk of what needs to be cooled will be in that area. Hell, even just a picture that shows where the actual die rests on the chip without the spreader would be enough. Vertical/horizontal probably indicates which 'direction' the rectangular die rests on the chip, latitudinal or longitudinal. And seeing as this is Haswell without a heatspreader, I can see where you'd want to follow along the line of the die: BIG HEADLINE fucked around with this message at 02:00 on Jul 4, 2015 |
# ? Jul 4, 2015 01:53 |
|
Isn't the point of a heat spreader to spread the heat? So you'll need to cool everywhere not just over the core.
|
# ? Jul 4, 2015 02:03 |
|
INTJ Mastermind posted:Isn't the point of a heat spreader to spread the heat? So you'll need to cool everywhere not just over the core. the point of a heatspreader is more to protect the chip.
|
# ? Jul 4, 2015 02:08 |
|
cat doter posted:I'm sure the difference between a horizontal and vertical pattern are very thermally significant For a non-square die, potentially yes. The PDFs linked on the Arctic Silver page actually show this pretty well by showing what the die looks like under the heatspreader and the area you want the compound to spread out over for ideal performance. If you do it opposite to the direction of the die you might not fully cover the important area. For a GPU or older/mobile CPU without a heatspreader patterns don't really come in to play, you want to actually spread the compound on the die as thinly as possible while still covering the entire thing. Thermal paste is only there to fill microscopic gaps, you don't need much of it to be effective as long as it covers the appropriate area.
|
# ? Jul 4, 2015 02:09 |
|
I need a new card and am on a roughly $200 budget. My GTX 760 is dying. Is this the best buck for my bang at that price? http://www.newegg.com/Product/Product.aspx?Item=N82E16814487133 Anyone have negative experiences with heat due to the size?
|
# ? Jul 4, 2015 03:19 |
|
Agreed posted:It wasn't that long ago that image quality mattered, like the generation before one of 'em (brain's telling me AMD) ripped off the other one's AF algorithm more or less entirely and it's been pretty neck and neck since then. Both of them have solid AA tech, no real difference there anymore apart from the details. Baseline AA is cheap and getting fancy isn't worth the cost at higher resolutions, especially now that monitors with higher DPI are coming out that really diminish the effects of aliasing.
|
# ? Jul 4, 2015 03:31 |
|
Manifest posted:I need a new card and am on a roughly $200 budget. Best bang for the buck is to save another $50 and grab a r290. A 960 is barely an upgrade at all.
|
# ? Jul 4, 2015 03:41 |
|
Manifest posted:I need a new card and am on a roughly $200 budget. Best bang for your buck at that price is to go with AMD. http://www.newegg.com/Product/Product.aspx?Item=N82E16814150705 http://www.newegg.com/Product/Product.aspx?Item=N82E16814125679 http://www.newegg.com/Product/Product.aspx?Item=N82E16814202099 http://www.newegg.com/Product/Product.aspx?Item=N82E16814121897R http://www.newegg.com/Product/Product.aspx?Item=N82E16814150706 td4guy fucked around with this message at 04:00 on Jul 4, 2015 |
# ? Jul 4, 2015 03:52 |
The Iron Rose posted:Best bang for the buck is to save another $50 and grab a r290. A 960 is barely an upgrade at all. Yeah, this one is probably your best choice, much faster than the lackluster 960 which can't even make good use of the 4GB of VRAM on the more expensive models.
|
|
# ? Jul 4, 2015 08:40 |
|
AVeryLargeRadish posted:Yeah, this one is probably your best choice, much faster than the lackluster 960 which can't even make good use of the 4GB of VRAM on the more expensive models. Then you're only ~$50 away from a 970
|
# ? Jul 4, 2015 09:24 |
|
Nevermind, it *was* 299.99. Probably will be again soon.
|
# ? Jul 4, 2015 10:16 |
|
dissss posted:Then you're only ~$50 away from a 970 The guy had a ~$200 budget and a R9 290 is already more than pushing it @ ~$260. A 970 would ~$100 over budget for similar performance at stock clocks.
|
# ? Jul 4, 2015 10:32 |
|
I think there should at least be a concerted effort to steer everyone away from the 960 because it's completely poo poo
|
# ? Jul 4, 2015 10:38 |
|
Anime Schoolgirl posted:I think there should at least be a concerted effort to steer everyone away from the 960 because it's completely poo poo The "960 is worthless" thing is really overstated. It is easily as good as a 380 (285).
|
# ? Jul 4, 2015 11:03 |
|
At 2GB the 960 is alright but it's not really seeing any improvements at 4GB because of the low memory bandwidth so it makes sense to pick the 380 every time.
|
# ? Jul 4, 2015 11:12 |
|
Riso posted:At 2GB the 960 is alright but it's not really seeing any improvements at 4GB because of the low memory bandwidth so it makes sense to pick the 380 every time. Hearsay. The 960 shows very real improvements from the 4GB framebuffer. Taking into account delta compression and the different clockspeeds, the real world bandwidth difference between the 960 and 285 is incredibly minor.
|
# ? Jul 4, 2015 12:13 |
|
The 280x is still an option around $200. Quick googles shows you can get them around 190 to 220 bucks after MIR. It handily beats the 969 from the benchmarks I've seen.
|
# ? Jul 4, 2015 14:00 |
|
|
# ? Jun 5, 2024 05:15 |
|
BurritoJustice posted:Hearsay. The 960 shows very real improvements from the 4GB framebuffer. Taking into account delta compression and the different clockspeeds, the real world bandwidth difference between the 960 and 285 is incredibly minor. http://www.guru3d.com/articles-pages/gigabyte-geforce-gtx-960-g1-gaming-4gb-review,1.html shows a massive difference of one fps in most games. Your link has like six fps at best. It's mostly margin of error. I am not sure I'd bother, with other more powerful cards around the same price point.
|
# ? Jul 4, 2015 14:11 |