Search results

It's that high because these are 99% two dies glued toghether with a DMI link/foveros. It's just unreasonable to put 40 fat x86 cores on a single die. So Intel's only option is to make smaller dies and link them. Since they don't have an energy efficient interconnect, you end up needing a ton of...

Tried updating to the latest chipset drivers with no success. Don't have any AsRock utilities..

Did anyone actually manage to put this to good use? Even tho in the documentation it says that my B450 and 2400G board and cpu are supported, the installer thinks otherwise: Also didn't the original version also support ram drives to speed up SSDs? Did they remove this in the new version?

If the difference in node pitch is so tiny, this just means that AMD's desktop designs are just this much better than Intel's desktop (power and performance per area), and that Intel is still ahead in laptops and possibly servers once IceLake-SP is released.

I need the Seagate FireCuda 120 SSD 1TB because I could sure use another ssd and 1 TB is a lot of storage!

Well darn, that explains it... The max for Raven Ridge is indeed 2133 with 4 SR sticks. I was looking at the manual from my phone when I was trying earlier and this is how the android reader displays it: Then I opened the exact same thing on the computer just now and the same table is...

@ThrashZone I'm using two kits of two memories, each is 2x8gb and they are suposedly the exact same brand and package. I know that's not ideal, I just thought that there might be a setting combination I'm missing. @Assimilator What I meant to say is, I tried 1.35V after you recommended it. I'll...

I didn't touch it, because I have no experience with this :D I just tried with 1.35V ddr voltage and +50mV NB voltage and everything else left on auto, and couldn't get it to post. It just gets stuck in a boot loop (memory training? ) and I powered it off after 3-4 minutes. I'll try re-seating...

I attached the complete html report as a zip. Here's what I get now in the Dram calculator: Should I be using the "Micron A-die" option? @theonek you're probably right, this could very well be the limit of the current setup. I was just hoping I could tweak the settings a bit and bring it...

@TheLostSwede this is what Thaiphoon Burner is showing me for all slots: The only thing that changes between them is the serial number and the manufacturing date, which seems logical. I see something about "8Gb A-Die (20nm) in there but not sure what's the equivalent in the dram calculator...

Hah, had the exact same problem when I originally set up the pc ~1.5 years ago with just two sticks - couldn't get it to boot reliably and be stable until I found some random reddit post that said to use slots A2 and B2 (also the mobo manual was absolutely horrendous). The problem now is that...

Hi, I can't seem to get the system to post at all using the stock settings. I'm using four sticks of 8gb, which are all the same brand and model (tho on closer inspection, the new pair has slightly different cooling case). The only way I managed to make the system boot is by reducing the ram...

Guru3d got about a 90W-ish delta between idle and and load, so if you add another 10W for base idle load, seems like it's really pulling around 95~100W max. https://www.guru3d.com/articles-pages/amd-ryzen-7-1800x-processor-review,7.html

Look at all that empty pcb space...

Ugh you're right, I'm way off - note to self don't do math when low on caffeine. Let's try this again. We compare performance relative to a single Skylake core. The BD core has half the IPC. When we multiply that by 40%, we'd still be at 0.5 * 1.4, -> 0.7. That would mean that in the best case...

Skylake's about twice as fast per core as Bulldozer, so even in the best case scenario where Zen is 40% faster, it'd still be 10% behind Skylake per core. That would of course still be perfectly fine as long as the price is right and you get to have more cores than an intel CPU in the same price...

The big die is actually the quad core CPU + 72 EU GPU and the small one is the L4 cache. They can't decouple the GPU from the CPU because they share resources... On the desktop side, the standard configuration comes with 24 EUs and takes roughly ~50% of the die. Therefore we can extrapolate...

Doesn't matter. As long as it's a laptop from a crummy OEM, the above still applies.

Dude seriously! You're blabbering about this in EVERY AMD driver thread. This is something that's beyond AMD's (or nVidia's) control . Most OEMs purposefully change the the ID and demand that the ID is not included in the standard driver stacks. That's because they're cheapscates and are too...

And you deducted this from what exactly? The AoS demo performs perfectly fine on GTX 900 and it supposedly uses AC extensively...

The tidbit from this news should be the fact that they're open-sourcing some of their libraries.

That CPU will be continue to be a fail, until AMD don't wake up and add turbo mode when just a single core is loaded. It should bump freq. to at least 2 GHz. Edit: Apparently the 6310 has turbo enabled...

I like that part but unfortunately, for the last couple of generations, the CPU portion of the die in Intel's CPUs has been getting smaller and smaller while the price per CPU has remained roughly the same (or has even increased) . And we'll keep getting screwed harder and harder without any...

I think that this is one of those rare occasions, where fanboys from the two camps should unite and collectively hate on Microsoft. When was the last time that Microsoft acquired a company and that ended up well for the company? Things don't look well for team Havok...

I'm just gonna leave this here: http://anandtech.com/show/9643/jim-keller-leaves-amd First they split the GFX division after all that hard work integrating it, now this. The little hope that I had left for Zen is evaporating quickly...