|
Zoom is bad on Windows and even worse on Linux
|
#
?
May 28, 2023 12:35
|
|
|
|
| # ? Jun 4, 2024 05:49 |
|
|
shrike82 posted:Gaming is bad on Windows and even worse on Linux
|
|
#
?
May 28, 2023 12:43
|
|
|
shrike82 posted:Zoom is bad on Windows and even worse on Linux We recently switched from Teams to Zoom. I still can't decide which is worse though.
|
|
#
?
May 28, 2023 12:43
|
|
|
idk i just use slack for everything
|
|
#
?
May 28, 2023 13:49
|
|
|
I was talking to a friend the other day about how I'd only just discovered Wallpaper Engine and how cool it was, when he tells me that he's had it for a while, but he doesn't use it anymore because his computer struggles enough already that even turning it on is a performance hit I was surprised because this friend of mine streams on Twitch pretty much every day, so I figured he must have had a decent PC. Turns out he was doing all of that out of a Zen 1 laptop so I invite him over to lunch today, and surprise him with a desktop build, with an i5-10400 and an RX 580 that I was no longer using. It's still an absolute workhorse of a card even after more than half a decade and I'm sure he'll get some good use out of it
|
|
#
?
May 28, 2023 13:57
|
|
|
Google Butt posted:idk i just use slack for everything Slacks great for text. It's got poo poo video call support though.
|
|
#
?
May 28, 2023 13:58
|
|
|
huddles are fine for ad-hoc stuff but you tend to need a zoom/teams-like especially for dealing with external parties
|
|
#
?
May 28, 2023 14:00
|
|
|
Discord is like, legit better than Slack in nearly every way apart from file upload support
|
|
#
?
May 28, 2023 15:07
|
|
|
BurritoJustice posted:The dual encoders are only on the 4070 and above, you get one on the 4060ti. It's still really fast, but not as crazy as higher end Ada. Huh, I completely missed that, I thought they always gave all cards the same decoder from top to bottom tier, in the past. Welp, in any event I just sniped an open-box 4070 on Best Buy nearby for $460, so if that works out I can forget about the 4060ti proposition.
|
|
#
?
May 28, 2023 15:35
|
|
|
I love this post but also 2014 was well past the point where OpenGL was clearly completely hosed and doomed, and you had big devs already looking for escape hatch. Remember AMD Mantle? This post is mostly about WebGPU, but has a nice history of OpenGL at the beginning: https://cohost.org/mcc/post/1406157-i-want-to-talk-about-webgpu quote:Everything was terrible! As it matured, OpenGL fractured into a variety of slightly different standards with varying degrees of cross-compatibility. OpenGL ES 2.0 was the same as OpenGL 3.3, somehow. WebGL 2.0 is very almost OpenGL ES 3.0 but not quite. Every attempt to resolve OpenGL's remaining early mistakes seemed to wind up duplicating the entire API as new functions with slightly different names and slightly different signatures. A big usability issue with OpenGL was even after the 2.0 rework it had a lot of shared global state, but the add-on systems that were supposed to resolve this (VAOs and VBOs) only wound up being even more global state you had to keep track of. Zero VGS posted:Welp, in any event I just sniped an open-box 4070 on Best Buy nearby for $460, so if that works out I can forget about the 4060ti proposition. Crysis suit voice: MAXIMUM VALUE
|
|
#
?
May 28, 2023 16:09
|
|
|
BurritoJustice posted:The dual encoders are only on the 4070 and above, you get one on the 4060ti. It's still really fast, but not as crazy as higher end Ada. https://developer.nvidia.com/video-encode-and-decode-gpu-support-matrix-new
|
|
#
?
May 28, 2023 16:16
|
|
|
4090FE shows as NLA on the  bestbuy.caNot that I've ever seen it available, either in store or shipped, on the bestbuy site since launch.
|
|
#
?
May 28, 2023 17:35
|
|
|
MikusR posted:4070 has only one encoder. 4070 ti has 2. Maybe a dumb question, but does it matter whether there's one or two if they lock all the consumer cards to 5 concurrent sessions?
|
|
#
?
May 28, 2023 17:36
|
|
|
Zero VGS posted:Maybe a dumb question, but does it matter whether there's one or two if they lock all the consumer cards to 5 concurrent sessions? Encoding latency, two distinct cores don't have to share the frame loads. Going more than two feeds, even on a two core card, work fine for archival/offline encoding but on live encoding you get bad lag from ingress to output.
|
|
#
?
May 28, 2023 17:44
|
|
|
slidebite posted:4090FE shows as NLA on the Yeah seeing Americans talk about how these cards are everywhere in this thread is wild as a Canadian. The only upside for me is that I wanted to upgrade my whole PC to get a second rig in the living room to play natively without streaming but since the 4090 is basically vaporware here, I never committed and I may have gotten over that idea since it's never available (aside from the really overpriced editions that are like double the inflated prices) when I checked several times a week for a few months. So that money is going to be used to rebuild the deck allowing me to enter my house.
|
|
#
?
May 28, 2023 18:03
|
|
|
Oh, the 4090 FE is definitely vaporware in the US—it's the AIB cards are plentiful. It seems like there are a number of AIB cards in stock in Canada too, so it looks like it's the same situation. The 4090 FE is $2100 CAD there it looks like, and the AIB cards start at around $2120. https://ca.pcpartpicker.com/products/video-card/#c=539&sort=price&page=1
|
|
#
?
May 28, 2023 18:08
|
|
|
Dr. Video Games 0031 posted:Oh, the 4090 FE is definitely vaporware in the US—it's the AIB cards are plentiful. It seems like there are a number of AIB cards in stock in Canada too, so it looks like it's the same situation. The 4090 FE is $2100 CAD there it looks like, and the AIB cards start at around $2120. https://ca.pcpartpicker.com/products/video-card/#c=539&sort=price&page=1 I was referring to third party cards too. None of the normal models were in stock when I was looking. Only a few exceptions where it's like 3000$Cnd+ and not even scalped. Newegg has nothing under 2600$ right now. Oh wait, what the gently caress. A lot of the "view details" instead of "Add to cart" sometimes means "yeah it's in stock and not the usual meaning of not in stock". I better stop looking further into this as I'm already on a 5800X CPU and 3080Ti so it's not like I really need it...
|
|
#
?
May 28, 2023 18:15
|
|
|
There are a ton of 4090s in stock for around $2100 right now and that's just looking at Canada Computers.
|
|
#
?
May 28, 2023 18:34
|
|
|
I don't think I will ever be "happy" about my 3060Ti FE purchase, but after reading some pretty savage 4060Ti reviews over the past few days, it certainly seems like I made the right choice by not trusting Nvidia to give a gently caress about budget builders. I would feel like even more of a fool if I had held out longer for such a questionable product.
|
|
#
?
May 28, 2023 19:29
|
|
|
It does take the sting out a little bit
|
|
#
?
May 28, 2023 19:33
|
|
|
shrike82 posted:why are you updating your work machine's kernel every couple months? I am using Ubuntu 22.04 and new kernels regularly come out and tank the computer. I ultimately don't update so frequently; I masked out the kernel packages and they are kept back. I don't remember what it was, but I was compelled to update some months ago, and while I finally had the normal procedure down, the initrd became so large that it refused to load. This is particularly stark considering I have drifted into kernel development in the past year and this stuff still punches me in the nuts. I've modified and built the 6.4 kernel more in the past three weeks than I ever touched a kernel from 1999 to last year. But normally I load that stuff in a VM or machine with a serial console. That is more manageable than booting into a rescue USB key, chrooting into my work computer's disk, and unfucking it whenever I update it's kernel.
|
|
#
?
May 28, 2023 20:17
|
|
|
Rocko Bonaparte posted:I've seen Nvidia's pervasiveness in software all over, but I can't help but remember the crushing pain I always feel when I have to update my Linux kernel. It's always a game of luck if dkms processes right and I have a display, or the united becomes so bug that I can't even use it anymore. It's a game I get to play with my work machine and it's NVidia GPU every few months. Otherwise, I get to sometimes reboot to regain one of my screens sometimes since nothing else works. I've used Fedora for five years, and never had a problem upgrading the kernel hundreds of times with the drivers from RPMFusion. If you're a performance beast who needs the very latest optimization drivers before those guys can package it, then that's on you.
|
|
#
?
May 28, 2023 21:06
|
|
|
At Mindfactory, 7600 and 4060 Ti both appear to have done about 1/10th of the numbers of the 4070 ti and 4070 on their launch days. Holy fuckballs that’s bad, low-end market is having none of it. To be fair everyone knows massive cuts are coming for 7600 over the next 2 months and the 4060 ti is not the one that people care about. So this particular launch cycle is stupid, nobody is excited about any of these and everyone paying attention is sitting out for another 2 months. https://www.notebookcheck.net/Lamen...0.720367.0.html
|
|
#
?
May 28, 2023 22:03
|
|
|
The problem with the 4060 Ti and 7600 is that the 3060 Ti and 6600 XT have already existed at these prices for months now. Nobody who was already unconvinced by those cards is suddenly going to want to buy one of the new ones. The discounts these last-gen midrange/low-end cards have been seeing only make the situation worse for the new cards.
Dr. Video Games 0031 fucked around with this message at 22:08 on May 28, 2023 |
|
#
?
May 28, 2023 22:06
|
|
|
BurritoJustice posted:I mean, the person paul is responding to is talking about their experience in 2013 as well. AMD did eventually rewrite their OpenGL driver, but it was literally only last year after it was broken for decades. The new one is faster but busted in it's own unique ways in that a bunch of things that used to work, don't anymore. If there is one area that AMD deserves a load of poo poo for, it's OpenGL. It’s unfortunately still true in a lot of ways. AMD has much better opencl support on paper, but a lot of their support is purely paper features and the actual functionality is broken in the runtime, or even just basic poo poo. Nvidia actually has the best opencl support in practice, because the features they claim to support actually work. And I don’t think things have improved that much on devrel either.
|
|
#
?
May 28, 2023 22:14
|
|
|
Paul MaudDib posted:At Mindfactory, 7600 and 4060 Ti both appear to have done about 1/10th of the numbers of the 4070 ti and 4070 on their launch days. Holy fuckballs that’s bad, low-end market is having none of it. At $400 street price the 4060ti is replacing the 3060ti with 3060 ti performance. It’s like cars the year there isn’t a new platform or a mid cycle refresh and the MY22 becomes the MY23
|
|
#
?
May 28, 2023 22:17
|
|
|
Craptacular! posted:I've used Fedora for five years, and never had a problem upgrading the kernel hundreds of times with the drivers from RPMFusion. If you're a performance beast who needs the very latest optimization drivers before those guys can package it, then that's on you. My work machine isn't running the 6.4 stuff I am doing. This is all happening with just apt using the official packages Ubuntu releases.
|
|
#
?
May 28, 2023 22:30
|
|
|
Hemish posted:I was referring to third party cards too. None of the normal models were in stock when I was looking. Only a few exceptions where it's like 3000$Cnd+ and not even scalped. Newegg has nothing under 2600$ right now.  I've seen $2100-2300 4090s in stock on Amazon for months now. Just never a FE on best buy.
|
|
#
?
May 29, 2023 01:21
|
|
|
Paul MaudDib posted:I have a very very heterodox opinion of what NVIDIA does. They've pretty consistently delivered innovation and dominance that follows, and that's what matters. I think NVIDIA is why graphics and GPUs can do all the crazy poo poo they do today. They push the envelope more than anyone else, even if they don’t always stick the landing. The Vendor B stuff reminds me of when we started doing WebGL and got to have a bunch of conversations with driver vendors about it. When we said to the big 3 (with about a year’s notice) that we wanted to put untrusted shaders through their pipeline, the responses were roughly:
Intel did a truly impressive job of fixing bugs as narrowly as possible, and not addressing adjacent bug surface until it was reported and escalated. We would have to report variants of the same bug a dozen times across different API entry points in order to get it fixed. It was like the “I’m not touching you, I’m touching your clothes” of kids in the back seat, but for the most widely deployed GPU platform in the world.
|
|
#
?
May 29, 2023 01:49
|
|
|
Craptacular! posted:I've used Fedora for five years, and never had a problem upgrading the kernel hundreds of times with the drivers from RPMFusion. If you're a performance beast who needs the very latest optimization drivers before those guys can package it, then that's on you. Yeah, I'm also running fedora with the prop drivers and secure boot, even, and haven't really had any issues, and I swear I get a new kernel every week or two. Had more issues with the amd apu drivers even though they're in the kernel although it seems they're finally sorted like a year after I got the laptop.
|
|
#
?
May 29, 2023 01:52
|
|
|
Subjunctive posted:I think NVIDIA is why graphics and GPUs can do all the crazy poo poo they do today. They push the envelope more than anyone else, even if they don’t always stick the landing. WebGL is a great little microcosm of how hosed OpenGL drivers are, it's an almost 1:1 port of OpenGL to the web and yet by the time the dust settled the browsers had all implemented it on top of DirectX instead
|
|
#
?
May 29, 2023 02:03
|
|
|
repiv posted:WebGL is a great little microcosm of how hosed OpenGL drivers are, it's an almost 1:1 port of OpenGL to the web and yet by the time the dust settled the browsers had all implemented it on top of DirectX instead That was the VERY clear advice from the vendors, so it’s not surprising but yeah it is pretty telling.
|
|
#
?
May 29, 2023 02:17
|
|
|
Subjunctive posted:I think NVIDIA is why graphics and GPUs can do all the crazy poo poo they do today. They push the envelope more than anyone else, even if they don’t always stick the landing. Did you get completely out of this space before the traditionally embedded silicon vendors got to relevant horsepower?
|
|
#
?
May 29, 2023 02:21
|
|
|
https://twitter.com/OrdinaryGamers/status/1662534956457885702 Looks pretty neat, Remix seems like a fun tool.
|
|
#
?
May 29, 2023 02:22
|
|
|
hobbesmaster posted:Did you get completely out of this space before the traditionally embedded silicon vendors got to relevant horsepower? My play (Firefox) had negligible-if-you’re-generous mobile footprint, so I didn’t really engage with Qualcomm or Imagination or whoever. Also I have never been paid enough to deal with Qualcomm, and I’m not sure it’s actually possible.
|
|
#
?
May 29, 2023 02:24
|
|
|
Subjunctive, I don't know how much I can say, but I very much feel that part about AMD and Intel in your post. Verrry much.
|
|
#
?
May 29, 2023 02:25
|
|
|
Cygni posted:https://twitter.com/OrdinaryGamers/status/1662534956457885702 found the source https://youtu.be/WxJyH-jK7mk Nvidia still hasn't released the official remix editing tools, but with the runtime portion being open sourced it seems some people are taking matters into their own hands
|
|
#
?
May 29, 2023 02:31
|
|
|
Subjunctive posted:My play (Firefox) had negligible-if-you’re-generous mobile footprint, so I didn’t really engage with Qualcomm or Imagination or whoever. Also I have never been paid enough to deal with Qualcomm, and I’m not sure it’s actually possible. Imagination/PowerVR means you have an extra layer of indirection when the silicon vendor drops a blob on you and everything is by necessity even more opaque and more finger pointing and delays occur. Qualcomm’s GPUs have an interesting history for anyone not familiar with them: https://en.m.wikipedia.org/wiki/Adreno
|
|
#
?
May 29, 2023 02:32
|
|
|
hobbesmaster posted:At $400 street price the 4060ti is replacing the 3060ti with 3060 ti performance. The 4060ti isn't great, but you don't need to exaggerate. There's certainly outliers where it's only as fast as a 3060ti, by on average it's on par with to slightly faster than a 3070.
|
|
#
?
May 29, 2023 03:56
|
|
|
|
| # ? Jun 4, 2024 05:49 |
|
|
https://twitter.com/PCMag/status/1663020622807732224
|
|
#
?
May 29, 2023 04:14
|
|