|
HalloKitty posted:Put it another way: would you think an overclocked Intel Core i7-6900K would struggle running 8 copies of a game that launched in 2004? Is it EQ2? Because if so, my answer would be "yes."
|
# ? Feb 27, 2019 15:52 |
|
|
# ? May 25, 2024 13:40 |
|
DrDork posted:Is it EQ2? Because if so, my answer would be "yes." ... OK, I guess you got me there
|
# ? Feb 27, 2019 16:05 |
|
HalloKitty posted:Of course it will work. A 2700X might have lower single-thread performance, but it certainly isn't lovely. Zen was comparable to Broadwell, and Zen+ is marginally better. On top of that the game has changed immensely from 2004. The original Vanilla minimum requirements were an 800 Mhz or higher CPU. The most recent expansion has an i5 760 as it's minimum.
|
# ? Feb 27, 2019 16:14 |
|
Otakufag posted:Would a 2700x be able to nicely multibox 8 accounts like that or it gets hampered by it's shittier single threaded perf? What I'm saying is, yes it can run 8 accounts no problem. Whether you want to go with a 2700x or 9900k is up to you. I'd expect the performance difference is similar to running 1 copy of the game, but I'm not very up to date on current WoW performance. I just know some idiot who five boxes with that setup I described above. Khorne fucked around with this message at 17:23 on Feb 27, 2019 |
# ? Feb 27, 2019 17:18 |
|
Intel released a video on their Lakefield thin and light SoCs with Foveros: https://www.youtube.com/watch?v=-besHp8HLxo It does look pretty interesting, especially if it will allow the little cores to run and do background tasks at the same time as the Sunny Cove core.
|
# ? Feb 27, 2019 19:57 |
|
I just watched the i9-9990XE video by der8auer, what a strange CPU. They essentially binned the very best i9-9980XEs and turned off four underperforming cores. It runs at 5.0 GHz Turbo with stock VID at 1.47V and all 14 cores bounce off the 110 Celsius limit three seconds into Cinebench using a (cold) 360mm custom loop. Seriously. 530W stock power consumption in Cinebench measured at the 12V connectors and the CPU is rated for 250W TDP. Zero overclocking headroom, in fact you’d have to clock it down to run it in a normal HEDT enthusiast system. The CPUs will be sold to OEMs only in an auction under NDA because they don’t trust regular consumers with cooling, setup and stability. They must be very scared of 7nm Ryzen benchmarks? The most interesting part is that Intel engineers insist that these voltages and temperature are perfectly safe for prolonged heavy use. eames fucked around with this message at 11:34 on Mar 4, 2019 |
# ? Mar 3, 2019 22:59 |
|
Why try to grasp at marginal increases in performance gains with multiple halo-tier products (edit: see also the W3175X) when merely just one will do? Is the frequency of harebrained product releases contributing to what all this "defending mindshare" talk is about? The only theory I can think of regarding what's going on behind the scenes is that Intel had multiple R&D teams try to come up with different ways to "max out" the 'Lake architecture and we got multiple plans rolling out almost at the same time. Sidesaddle Cavalry fucked around with this message at 23:25 on Mar 3, 2019 |
# ? Mar 3, 2019 23:22 |
|
Not gonna lie, watching Intel squirm like this is kinda amusing, even after all this time. Can’t wait for the I7-9999XE where it crashes under any benchmark level loading but is totally stable and fine for daily desktop use so long as that only involves email and word processing and Intel only sells it under NDA to the admin secretary market.
|
# ? Mar 4, 2019 00:43 |
|
Can't wait to see what happens when AMD releases their new Threadrippers, with the go-to middleground model being 32 cores and say 4.5GHz+ boost.
|
# ? Mar 4, 2019 01:15 |
|
My favorite part about those new HEWS cpu's or whatever market segment they are going to create/pretend they fit into is the socket itself. Every reviewer I've seen has complained about the chip not locking into the board with anything other than the pressure of the cooler and when you remove the cooler the cpu just kind of comes out/flops around. Is that standard operating procedure for those types of chips/boards? In Linus' video where he unboxed a system that came pre-built from Intel themselves, it already had a gaggle of bent pin's prior to ever even turning the system on.
|
# ? Mar 4, 2019 01:36 |
|
In 1U computers where millimeters matter it makes sense, but it seems strange they didn't redesign the retention system for desktop use. They already do it for the same CPU in laptops and desktops, the needs of thin rack mount systems are just as unique.
|
# ? Mar 4, 2019 10:38 |
|
I fixed my old t42 laptop’s dead cfl backlight with the community made led replacement, and the pentium m isn’t ALL bad, imho?? it’s still not good but decoding video in hardware is for dorkos. and the think light has always been tops over backlit keys. a laptop with both top and bottom illumination, the brightness of each independently adjustable to the point of being off, is the end game so probably too good for this world
|
# ? Mar 4, 2019 11:23 |
|
if it does exist I just assume the manufacturer decided to use a think nub pointer but somehow made it backlit with a laser that uses he laptops webcam to track the users eyes so that it can always blinds the compute toucher
|
# ? Mar 4, 2019 11:25 |
|
Holy Jesus of Nazareth, Thunderbolt 3 is going royalty free. Problem is, does this do anything actually given other costs plus the problems with USB 3.1+ and USB-C among manufacturers? https://arstechnica.com/gadgets/2019/03/thunderbolt-3-becomes-usb4-as-intels-interconnect-goes-royalty-free/?amp=1
|
# ? Mar 4, 2019 18:07 |
|
Can't wait for USB 3.3 Gen2x2x1
|
# ? Mar 4, 2019 18:12 |
|
USB 3.e^-i pi Hamiltonian(3^theta)
|
# ? Mar 4, 2019 18:22 |
|
OK, dumb question, with USB-C 3.1+, why should we still care about Thunderbolt? Are there even enough devices on the market to care?
|
# ? Mar 4, 2019 18:44 |
|
.
sincx fucked around with this message at 05:55 on Mar 23, 2021 |
# ? Mar 4, 2019 18:47 |
|
I suppose I should have clicked on and read the article, since it talks about USB4.
|
# ? Mar 4, 2019 18:53 |
|
Combat Pretzel posted:OK, dumb question, with USB-C 3.1+, why should we still care about Thunderbolt? Are there even enough devices on the market to care? thunderbolt is still the only game in town if you want egpu. being able to pass through native pcie makes it much better for docking stations including chaining displays
|
# ? Mar 4, 2019 19:27 |
|
sincx posted:I'll take "every port and cable have the same features and specifications" as a huge win. Cheap usb dogs dogshit vendors aren't going to pay for pcie
|
# ? Mar 4, 2019 19:36 |
|
Also, cheap USB-C cables aren't going to vanish, either.
|
# ? Mar 4, 2019 20:39 |
|
This looked like a huge win for computing when I first read it (40gbps pcie ports in all phones and tablets! yay!) but then I quickly realized it'll probably lead to a mess of ports with the same name, same look and completely different functionality. hope I'm wrong.
eames fucked around with this message at 22:32 on Mar 4, 2019 |
# ? Mar 4, 2019 20:50 |
|
eames posted:This looked like a huge win for computing when I first read it (40gbps pcie ports in all phones and tables! yay!) but then I quickly realized it'll probably lead to a mess of ports with the same name, same look and completely different functionality. hope I'm wrong. This is exactly what will happen.
|
# ? Mar 4, 2019 20:59 |
|
NeuralSpark posted:This is exactly what will happen. It is exactly what has happened. USB-C is (or can be) used as the physical interface for: USB 2.0, 3.0, 3.1, 3.1 GEN 2, USB 3.1 GEN 2 SSGSS Thunderbolt 3 VirtualLink Displayport Alt mode MHL Alt Mode HDMI Alt Mode Audio Adapter Accessory Mode On the plus side, I think every port supports USB 2.0 as a minimum, and usually support 3.1 if they support any of the Alt modes.
|
# ? Mar 4, 2019 21:26 |
|
those new USB names are impressively bad Ice Lake having TB3 on package is Good tho
|
# ? Mar 4, 2019 21:26 |
|
Lowen SoDium posted:It is exactly what has happened. USB 3.0 is USB 3.1 Gen1/USB 3.2 Gen1. Also, everything is going to remain a huge mess and getting a cable that supports what you're interested in needlessly tedious. Lambert fucked around with this message at 21:32 on Mar 4, 2019 |
# ? Mar 4, 2019 21:30 |
|
Lambert posted:USB 3.0 is USB 3.1 Gen1/USB 3.2 Gen1. USB 3.0 can also be called USB 3.2 Gen 1x1 as well, and also marketed as USB SuperSpeed. This is not to be confused with USB 3.1 Gen 2 aka USB 3.2 Gen 2x1 or USB 3.2 Gen 2 aka SuperSpeed+ But SuperSpeed+ can also be used to market USB 3.2 Gen 1x2 and USB 3.2 Gen 2x2 that are actually 20gb instead of 10gb (only 2x2 multi link mode, obviously you dumb consumer) How could ANYONE get these confused, its clear as day!!! Cygni fucked around with this message at 21:44 on Mar 4, 2019 |
# ? Mar 4, 2019 21:39 |
|
It's a USB standard, so it's not going to be right on the first try. They'll have to flip it once (or twice) before it sticks
|
# ? Mar 4, 2019 21:51 |
|
I never had a problem inserting USB devices into USB ports in the dark, but my case has USB and Firewire ports right next to each other, Try fitting a USB device into a Firewire port for like a minute to really know confusion.
|
# ? Mar 4, 2019 22:01 |
|
.
sincx fucked around with this message at 05:55 on Mar 23, 2021 |
# ? Mar 4, 2019 22:54 |
|
Thanks Intel for letting go finally but, uh... This is getting fixed, right...? Yes it's called the Thunderclap
|
# ? Mar 5, 2019 03:30 |
|
canyoneer posted:It's a USB standard, so it's not going to be right on the first try. They'll have to flip it once (or twice) before it sticks
|
# ? Mar 5, 2019 08:54 |
|
Malcolm XML posted:LOL never gonna happen This is going to happen if you're strictly limiting yourself to the USB 4.0 label (that's the point after all), but since it's also backwards compatible it won't stop the mess that is USB 3.x and USB-C connectors from existing, so instead, the dogshit vendors are just going to keep selling "USB 3.x" cables that lack power delivery or any of the display modes or whatnot.
|
# ? Mar 5, 2019 09:11 |
|
https://www.theregister.co.uk/2019/03/05/spoiler_intel_flaw/
|
# ? Mar 5, 2019 10:19 |
|
Here are the money quotes from that article:quote:The researchers also examined Arm and AMD processor cores, but found they did not exhibit similar behavior. Looks like Intel just found out about it in Dec 2018 so Icelake probably won't have hardware mitigations for this thing either. edit: \/\/\/\/\/\/\/ They really seem to love the artificial product segmentation it allows though. Maybe if they get desperate to get something out if there is a big wave of successful attacks using this method and no other fix is available in a reasonable time frame I can see them going that route. PC LOAD LETTER fucked around with this message at 13:27 on Mar 5, 2019 |
# ? Mar 5, 2019 13:00 |
|
Since this is somehow related to Rowhammer maybe this will force them to enable ECC on consumer platforms?
|
# ? Mar 5, 2019 13:23 |
|
I sure as gently caress hope so, to maybe finally get higher speed modules, that also happen to be affordable.
|
# ? Mar 5, 2019 13:31 |
|
i don't remember rowhammer attacks taking weeks, especially javascript-based. guess those pocs didn't exist to make the paper sound better architecture attacks not being universal isn't revolutionary either, especially when they're focusing on one vendor
|
# ? Mar 5, 2019 14:30 |
|
|
# ? May 25, 2024 13:40 |
|
Wiggly Wayne DDS posted:i don't remember rowhammer attacks taking weeks, especially javascript-based. guess those pocs didn't exist to make the paper sound better Wiggly Wayne DDS posted:especially when they're focusing on one vendor quote:The researchers also examined Arm and AMD processor cores
|
# ? Mar 5, 2019 14:42 |