|
Probably not. This is also looking like a smoke and mirrors operation to distract from his insider trading some time ago when he sold off shares just before Intel announced the Spectre and Meltdown vulnerabilities.
|
# ? Jun 21, 2018 16:36 |
|
|
# ? Apr 24, 2024 16:40 |
|
SourKraut posted:Wonder if he’d still have been retained if they were doing better on the CPU side. It could be a face saving move, an affair is much less embarrassing than losing a 5 year lead in 5 years.
|
# ? Jun 21, 2018 16:37 |
|
Mark Larson posted:Probably not. This is also looking like a smoke and mirrors operation to distract from his insider trading
|
# ? Jun 21, 2018 16:43 |
|
Mark Larson posted:Probably not. This is also looking like a smoke and mirrors operation to distract from his insider trading some time ago when he sold off shares just before Intel announced the Spectre and Meltdown vulnerabilities. "drat they can get me for securities fraud...better stick my dick into an employee."
|
# ? Jun 21, 2018 16:43 |
|
Seamonster posted:"drat they can get me for securities fraud...better stick my dick into an employee." Works for me
|
# ? Jun 21, 2018 16:56 |
|
Any port in a storm.
|
# ? Jun 21, 2018 17:10 |
|
....how hosed do things have to be that sticking your dick in someone gets you a vastly quicker exit than securities fraud? One would think they would be equally quick, or the securities fraud only a little behind!
|
# ? Jun 21, 2018 17:11 |
|
SwissArmyDruid posted:....how hosed do things have to be that sticking your dick in someone gets you a vastly quicker exit than securities fraud? I don't know, but I certainly intend to find out!
|
# ? Jun 21, 2018 17:21 |
|
Mark Larson posted:Probably not. This is also looking like a smoke and mirrors operation to distract from his insider trading some time ago when he sold off shares just before Intel announced the Spectre and Meltdown vulnerabilities. SwissArmyDruid posted:....how hosed do things have to be that sticking your dick in someone gets you a vastly quicker exit than securities fraud? Whether the government will step in and say "hey, it's not okay THIS time" even though they didn't the past few thousand times, who knows. Khorne fucked around with this message at 18:24 on Jun 21, 2018 |
# ? Jun 21, 2018 18:19 |
|
New AMD promotion: turn in your CEO, we'll exchange it for a free Raja
|
# ? Jun 21, 2018 18:48 |
|
How big is the new Intel socket going to have to be to handle Rajas shroud?
|
# ? Jun 21, 2018 18:51 |
|
Subjunctive posted:Any port in a storm. I didn't know we were up to USB Type D
|
# ? Jun 21, 2018 23:04 |
|
NewFatMike posted:I didn't know we were up to USB Type D We already have A and C to insert our plugs into...
|
# ? Jun 22, 2018 15:36 |
|
HalloKitty posted:We already have A and C to insert our plugs into... Heh.
|
# ? Jun 22, 2018 16:43 |
|
So um https://www.youtube.com/watch?v=AF_fRowqscM That sure is a thing.
|
# ? Jun 23, 2018 12:04 |
|
Do all Samsung B-die modules clock fast, or are these chips also binned? I've been looking for fast ECC DIMMs, and I've stumbled across ones from Samsung themselves having their B-dies. But the fastest DIMM of theirs is DDR4-2400 CL17, a far cry from DDR4-3200 CL14.
|
# ? Jun 23, 2018 12:55 |
|
It's not a binning process, it's just validation of primary, secondary and tertiary memory timings. It just so happens that Samsung B-die can use some very aggressive timings with AMD Ryzen parts. When you drop random memory kits into a motherboard, the motherboard checks its internal database to see if there is a validated memory timing that the motherboard vendor has determined to be "works best". If there is a validated memory timing, the motherboard loads those timings. We find out if the motherboard has those timings by looking at the memory QVL. If there is not a validated memory timing, the motherboard runs through algorithms and plugging in random values to obtain memory timings stable enough to boot. This is what happens with memory training. As for ECC, the fact of the matter is that ECC will not overclock as easily or as high as non-ECC memory. ECC is made for stability. SwissArmyDruid fucked around with this message at 13:47 on Jun 23, 2018 |
# ? Jun 23, 2018 13:43 |
|
Combat Pretzel posted:Do all Samsung B-die modules clock fast, or are these chips also binned? I've been looking for fast ECC DIMMs, and I've stumbled across ones from Samsung themselves having their B-dies. But the fastest DIMM of theirs is DDR4-2400 CL17, a far cry from DDR4-3200 CL14. Khorne fucked around with this message at 14:29 on Jun 23, 2018 |
# ? Jun 23, 2018 14:27 |
|
Khorne posted:There are some 16gb ecc b-die sticks that will stable oc to 2933 / 16? (or maybe 17). OCing with ECC RAM is kinda fun because you see a fountain of errors when you get too aggressive. It's loving stupid that ECC is not standard at this point given the existence of Rowhammer exploits that can literally be run from the web. On top of that, yes, super useful for OC'ing because the memory will literally tell you when it's starting to poo poo itself. TR and certain pre-Broadwell Xeons have a unique trait here.
|
# ? Jun 24, 2018 07:03 |
|
TheCoach posted:So um Okay, that's it who the gently caress in this thread works for AMD Marketing? I'm calling you out.
|
# ? Jun 24, 2018 07:33 |
|
FaustianQ posted:Okay, that's it who the gently caress in this thread works for AMD Marketing? I'm calling you out. 3/10, no "monster truck of computing", are you even trying bro
|
# ? Jun 24, 2018 07:35 |
|
Paul MaudDib posted:
The boost algorithms (iirc XFR2 communicates with the motherboard to get information about VRM capabilities) and hidden on-die sensors are already good enough that you'd have to be crazy to manually overclock a TR2 CPU other than for the sake of benchmark highscores. There may be people out there that would trade an additional 15% performance boost spread over 64 threads for a 150% power consumption increase but I can't imagine that there are many of them. The whole point of these super high core count CPUs is to run them at or below their perf/watt sweet spot, of course they're going to produce dumb numbers if you apply 6-8C standards to them, see Intel's 5 GHz chiller demo. speaking of internet drama, I worry about adoredtv's health if this isn't a hoax: https://babeltechreviews.com/nvidias-ceo-is-the-uncle-of-amds-ceo/
|
# ? Jun 24, 2018 10:03 |
|
This power phase bullshit pretty much just applies to the 24C and 32C variants, right? Who the gently caress cares about these? Only professionals and PCMR retards will realistically get those, former because of need and latter because of bragging rights. And only latter will overclock these, so not sure why anyone cares. Price will prevent prosumers from buying them. The 32C one tickles me for all the casual VM and compiling bullshit, but as said, price is a barrier. Also, the double bandwidth between dies goes away with that one. Which I assume will have an effect on inter-die latencies.
|
# ? Jun 24, 2018 10:21 |
|
Combat Pretzel posted:This power phase bullshit pretty much just applies to the 24C and 32C variants, right? Who the gently caress cares about these? Only professionals and PCMR retards will realistically get those, former because of need and latter because of bragging rights. And only latter will overclock these, so not sure why anyone cares. Price will prevent prosumers from buying them. Do we know what it will cost? I am tempted to switch my 16C for the 32C version.
|
# ? Jun 24, 2018 11:11 |
|
Not yet. Probably twice the 16C version. Maybe a little less, either as a gently caress you to Intel, or because there's a chance the bins on it might be a little worse, with what the official base and boost clocks being lower than the 16C version.
|
# ? Jun 24, 2018 11:35 |
|
Paul MaudDib posted:(realistically they will probably just throttle on existing boards, not melt the board. Existing boards will probably be limited to 1.2 volts or less when running all-core loads under air, which is probably... 3.5 GHz? And less under AVX.) Do Zen chips actually have appreciably higher power use on AVX loads? I figured the power draw would be similar to SSE since Zen has half-rate AVX ops.
|
# ? Jun 24, 2018 13:54 |
|
Combat Pretzel posted:Not yet. Probably twice the 16C version. Maybe a little less, either as a gently caress you to Intel, or because there's a chance the bins on it might be a little worse, with what the official base and boost clocks being lower than the 16C version. Hmmm, that will probably be around 2000 Euros, that's a bit much.
|
# ? Jun 24, 2018 15:17 |
|
Mr Shiny Pants posted:Do we know what it will cost? I am tempted to switch my 16C for the 32C version. my prediction 32c - $1600 24c - $1200 16c - $800
|
# ? Jun 24, 2018 16:37 |
|
repiv posted:Do Zen chips actually have appreciably higher power use on AVX loads? I figured the power draw would be similar to SSE since Zen has half-rate AVX ops.
|
# ? Jun 24, 2018 19:07 |
|
Any last-minute advice before I attempt to fix a few crooked pins on my Ryzen CPU? I値l be very careful, but more than that is needed when a 100% success rate is required. I値l be wiggling a toiny flat iFixit screwdriver to bring them upright, then I値l use a rolled out boxcutter to align them properly. Still don稚 know whether, when nor how they might break though.
|
# ? Jun 24, 2018 22:45 |
|
ufarn posted:Any last-minute advice before I attempt to fix a few crooked pins on my Ryzen CPU? I値l be very careful, but more than that is needed when a 100% success rate is required. https://www.youtube.com/watch?v=y8U2NkbiMAI
|
# ? Jun 24, 2018 22:57 |
|
I was hoping that would involve a potato and a plunger or something.
|
# ? Jun 24, 2018 22:59 |
|
Did you open palm slam it into the socket or something? I won't be laughing when I do this inevitably. GRINDCORE MEGGIDO fucked around with this message at 01:19 on Jun 25, 2018 |
# ? Jun 24, 2018 23:16 |
|
I really hope AM5 has a better retention mechanism. Nothing wrong with PGA, just there really shouldn't be a risk of damage when just removing the cooler.
|
# ? Jun 25, 2018 00:11 |
|
It would be nice if there was a frame that swung down over the top like LGA. Doesn't have to be a precisely tensioned doohickey like LGA, since the ZIF socket makes contact with pins, just something to keep it from getting yanked out by the cooler.
|
# ? Jun 25, 2018 01:03 |
|
I just moved back from Europe to Australia, but I managed to leave the part of my motherboard that the cooler screws onto behind. I've currently got my computer set up open air on top of the motherboard box with the cooler simply held down by gravity. The temperatures are actually quite good, hitting about 30C at idle. Should I even bother asking ASRock if they can send me a replacement? What other options are there?
|
# ? Jun 25, 2018 01:26 |
|
What if Hector Ruiz became the next Intel CEO
|
# ? Jun 25, 2018 01:28 |
|
Measly Twerp posted:I just moved back from Europe to Australia, but I managed to leave the part of my motherboard that the cooler screws onto behind. I've currently got my computer set up open air on top of the motherboard box with the cooler simply held down by gravity. The temperatures are actually quite good, hitting about 30C at idle. My guess is that ASRock will send you the part for a small cost, plus silly shipping. Likely still cheaper than a new cooler.
|
# ? Jun 25, 2018 01:29 |
|
|
# ? Apr 24, 2024 16:40 |
|
Bloody Antlers posted:What if Hector Ruiz became the next Intel CEO Nah, Tom Forsyth.
|
# ? Jun 25, 2018 01:30 |