Jump to content

jasonvp

Member
  • Posts

    1,019
  • Joined

  • Last visited

Everything posted by jasonvp

  1. The thing with Wikipedia is: anyone can submit articles to it. Here's an interesting tidbit: MR fluids have been used in vehicle shock absorbers in GM high performance vehicles since the late 90s (experimental) and early 2000s (production). That's a pretty stout "commercial application" if ever there was one. Nothing you're going to do with your thumbs to a joystick over the course of its short life is going to match the stress put on shock absorbers in a Corvette or high performance Cadillac (or any of the other sports cars that have adopted the tech from GM).
  2. Yes. Bait. And they don't have a "bad reputation"; not sure where you're getting that unless you're specifically referring to certain GPUs. BTW: the problem isn't actually affecting their sales, so... so much for said reputation? In the networking world, the purchase of Mellanox in combination with Cumulus is making for a massively powerful SmartNIC product of the likes the industry hasn't seen before. Look to the Bluefield-2 and Bluefield-3 DPUs. They're.... actually fucking awesome. And only really possibly given Team Green's oversight.
  3. As the grumpy ol' John Dvorak once wrote: He wrote that more than a decade or so ago, and he's been right all along. IB's got a great foothold because of its insane low latency, which is fantastic specifically for HPC workloads. But there are tricks that can be done with Ethernet to rapidly approach IB's low-latent links.
  4. That's why I put quotes around the word standards.
  5. Given all the crap that Asus went through a month or so ago with the X3D series chips and melting, I decided to, for the first time in forever, not use them. My new rig is built from an MSI 670E Godlike motherboard which is supposedly E-ATX. The case in question is one of the (not the) largest in existence: the CaseLabs THW10. Its motherboard tray is large enough to easily fit most E-ATX sized boards. Except the Godlike, apparently. Two pics. The first is looking into the case where you can see along the left side that the motherboard tray can't line up properly with the rear of the case. And in the center of the pic you'll see that the tray isn't fully pushed in. It simply can't be. Second pic is behind the case, showing the tray hanging out by about an inch or so: I have a zip tie in place to keep the tray from moving and putting pressure on the water lines. That could turn out disastrous... Good thing there are "standards" for motherboards, eh?
  6. Cadillac CT5-V Blackwing. 670HP, six-speed manual transmission, and just plain comical to drive.
  7. Blackwinnnnnnnng! Just picked it up last Friday. Still working on the 500 mile break-in before I crack 'er open.
  8. I'll bet money they'll see a sharp decline in the user base if they move to a sub model for the Mac software. And Blackmagic will see another surge of new users, all on Macs this time. I'm sure the BM guys are all rubbing their hands together, anxiously waiting for Apple to shoot themselves in both feet.
  9. That's because Twitch isn't allowed to use all of Uncle Jeff's entire infrastructure to universally transcode. Youtube's streaming is allowed to use any part of GOOG's impressive infrastructure it needs. Live, real time transcoding for all resolutions up to and including the one streamed. For anyone and everyone. And at a significantly higher input bit rate limit of 51Mbit/sec. From a video perspective, Youtube streaming has been eons beyond anything Twitch has produced for quite some time.
  10. Of no real relevance to the story but, Malta?! I'd no idea GF had built a plant in Malta. That's the area of NY that I grew up in. Er. Wait. Maybe "grew older" in. I've yet to grow up.
  11. Video by Nvidia, cooling by EK, and power delivery by CableMod.
  12. For better, for worse, and from my experience with the 5800X3D, it really depends on the game. Poorly optimized games that are CPU-bound seem to really thrive with the 3D cache. Example: Star Citizen, which I play a LOT of. Once I dropped that 5800X3D in, performance in that game skyrocketed. Which is sad, but expected. However, pivot over to the new CoD MW2 title, and that's GPU-bound for me. I play 4K, ultra (including AA and AO), without DLSS. I have a 140FPS soft-cap to stay within the G-Sync range of my display, and the game usually stays right there at 140. The GPU is at or near 95-98% use during the gaming because Reflex is keeping it from hitting 100%. So, with something like CoD, the 3D cache doesn't matter as much. The benchmarks from the reviewers will be key in making the decision. I am definitely getting one of the three of them. But I'm really hoping the reviewers take some time and specifically focus on the "bad" games that the previous 5800X3D made such a difference with. Test each of those games with each of the new processors. My gut tells me that, assuming Windows can keep everything scheduled properly (haaaahahahahahahaha *cough* *sputter* sorry there. Hahahahahahahaa...) the 7950X3D will be the bruiser of the group.
  13. The OP is sort of conflating what happened. I think it's safe to say that the terminal folks use on any *NIX system got its genesis from X11. Sure, VT100 ANSI-based terminals existed far longer than X11 did, but what we're used to playing on and with is a graphical representation of said. And that came from X11. Each *NIX OS had its own X11 server, and DEC was certainly no stranger to it. But that far back, it was their Ultrix OS, not the Alpha and OSF/1 (original name for the UNIX that ran on the Alpha). Ultrix existed way, way before the Alpha did, but it still had ye ol' X11 server and the xterm application. Make no mistake: the Alpha was awesome for its time. I attended Clarkson University from '91-95 and the school was saturated in IBM RS/6000 workstations and servers. They were pretty fast; certainly trounced the snot out of the Sun workstations and servers we had. The last year I was there, we finally got an Alpha and it was eye opening. One of the guys I worked with there had written a Fortran 77 program that computed pi out to some ungodly number of digits. The quickest time he could get out of the fastest RS/6000 we had was almost 21 minutes. The Alpha did it in under 7. A couple of years after I graduated I found myself as a fledgling network engineer at a little ISP no one's ever heard of: AOL. We had fleets of Alphas used as web proxies for our dial-up users. The I/O throughput on them was insane and the 64-bit memory addressing helped keep all that stuff in a RAM cache for quick retrieval. I even had one under my desk as my workstation.
  14. It appears to be a bit more than that. Or, rather, it's a bit more involved. The issues seems to be production variances in the adapter cable itself. See Jay's latest video. His was soldered and built quite well. The one Igor had looked like a 5 year old soldered it. Either way I agree with the first part of your statement: invest in another cable from Cablemod and run it all the way back to your PSU. If space is tight, get the 90* adapter from them, too. Just to be one the safe side.
  15. I've heard stories of folks getting AIBs. Unfortunately, NVidia decided (again) to lean on BestBuy in the US to sell their FE cards. BestBuy's site alternates between "Sold out" and "Sold out, and we don't ship this product". The depletion happened in less than 5 seconds, but it's still not clear whether BestBuy will actually ship, or require you to pick up. Either way, I'll bet we see a lot of these on Ebay later today. Embrace the scalp. It's going to happen again.
  16. I'm buried in meetings all morning and haven't had a moment to read up or watch the vid: has NVidia said if the 4090 is using 1 or 2GB chips for the VRAM? IOW: will the RAM be on one side of the card or both?
  17. I'm pretty sure he wasn't talking about modifying anything. MSI Afterburner is a thing, and let's you dial up or down the power usage of the card.
  18. If this actually happens (note the emphasis on IF) I can see a migration over to YouTube. The barrier to entry for monetization on YouTube is somewhat high (1000 subs, certain number of hours watched, etc) but a larger streamer could pretty easily cross that and begin making money. Once they do cross that threshold, the sky is literally the limit. The streamer UI on YouTube is atrocious (they know that) and pales in comparison to Twitch's. The chat on YouTube isn't as feature-rich as Twitch's. And there are some streamer features on Twitch such as gifted subs, etc, that aren't yet available on YouTube but are supposedly coming. The "bot API" for lack of a better phrase is really slow and done in batches, I think. That means if someone in chat fires off one of your !commands for a bot to answer (eg: "!specs" and the bot responds with your PC specs) it can take a couple of seconds before the bot responds. That needs to be looked into. The benefits? EVERYONE gets transcoded to ALL resolutions from your base streamed resolution down to 360p (maybe lower?) It doesn't matter what status you are as far as monetized, partner, etc. EVERYONE has immediate access to the transcoders. And they work wonderfully and quickly. YouTube also has a significantly higher max bitrate; the official one is 51Mbits/sec. Unofficially? ... Technically, and possibly financially, YouTube is a vastly superior streaming platform. From a community perspective, it's still "work in progress."
  19. I don't like the some of Elon's product ideas or how he's gone about getting them mainstreamed. However, I do think he's genuinely a good person and he truly means well for the human race. Whether you agree with him or not is another story, of course. But I think he sees how much of a cesspool Twitter has become, specially with the way it takes control of what it considers good/bad speech. I think this is a good thing for Twitter.
  20. And is there an issue with that? As has been pointed out by a few in this thread: the 3090Ti is for the person who's pursuing performance at any cost, not some made-up $/performance metric. If you're focused on the latter, the 3090(Ti) is the wrong card for you.
  21. As I've explained on LTT a few times in the past: this is NVidia following the exact, same pattern they do with every previous generation. The year they intend to change generations (eg: this summer), they also release a top of the top, best of the best version of the current generation. That, and only that, is the reason this product exists. They did it in the 2000-series, the 1000-series, the 900-series, etc, etc. And they'll do it for the 4000 series, too. On and on. It's their usual cadence.
  22. I could be mistaken, but I believe this is the first full bandwidth Thunderbolt [3|4] cable at that length. The previous "hard" limit was 2M.
  23. And when we can get all those different types of data streams, concurrently, over an HDMI 2.1 cable, lemme know.
  24. Bitch all you want about the price. I'll betcha that when the rest of the manufacturers catch up with their own 10ft Thunderbolt-4 cables, they'll all cost about the same. Maybe slightly less expensive, but the money saved won't buy you a cup of sludge from Starbucks. Thunderbolt cables are fucking expensive, kiddies. That's just how it goes. Being able to carry 40Gbits/sec in a non-fiber cable ain't cheap.
×