Yup. Hope that's sorted. Still have a bit over a week before I look at my build and well, the C6H is the only decent board available in the country.
Plenty of stock still but I'm not buying a faulty board
Things that make you go hmmm.
Do you think foul play? I wouldn't be surprised if this was an underhanded tactic to impact ryzens effect on the market. But this of course is borderline tinfoil hat stuff
its simply new platform, similar stuff happens when intel pushes something completely new. (like when they released lga2011 for first time, or any other socket; it wasn't noticeable with sandy-skylake, as nothing really changed there; in here they have something completely new on their hands.)
Fair enough. I do think I made the right decision going for kaby Lake. Ryzen is a new and immature system and I would rather not have to h go through the whole "working out the kinks" phase for my first build. I will be looking forward to their future tech especially when multi core apps and games become the norm.
You will be fine with kabylake
I am stuck with my current pc for the foreseeable future but I definitely intend on upgrading to a multicore monster next time around...
4 Likes
does anyone know how Ryzen is implementing VT-d if at all? I am looking at android studio development and true VT-d is required under Windows to run the anroid VM's in android studio.
amd has its own 'amd-v' tech. Its widely recognized.
In case anyone is wondering, ECC is supported (board maker has to support it and it's UDIMM only)
AM4 socket is going to be around for a while. If it's like previous sockets a BIOS update should be all that's needed to support new chips.
Zen2 and Zen3 are going to be a thing.
ITX boards are up to manufacturers. Also, APUs coming soon(tm)
All swiped from the recent Reddit AMA.
1 Like
I don't think this video has been linked yet:
Buildzoid is a components guy and a ln2 overclocker. He does some PCB breakdowns for GamersNexus now and then.
RAM is interesting, he lists the memory multipliers Ryzen support stock, it's just four and the highest x32 is a bit flakey according to reviewers. He talks about the new RAM kits form G-Skill. How you need a top end motherboard with an external clock-generator to use a diff base clock. Otherwise your PCIe, SATA etc will have a bad time.
Good video with many details.
1 Like
This is an interesting article:
So you would expect a Bell-curve. When two humps or spikes appear, that could point to issues. Can we expect a bell curve on the hoRyzen? If the windows issues are fixed, a patch and firmware update here and there, you'd think so.
2 Likes
Ryan is overclocking the 1700 (none X) using air cooling:
https://www.pcper.com/reviews/Processors/Overclocking-AMD-Ryzen-7-1700-Real-Winner
He gets his sample up to 4 GHz, a very good overclock. Especially nice with air cooling. Performance increase as expected. Better than stock 1800X as that one tries to keep power draw under control. Power consumption for that overclock increases the most though, over 100 Watts more.
About Vcore:
I have been having conversations with AMD, ASUS and other motherboard vendors on the reporting of voltages and temperatures for Ryzen processors (in addition to even clock speed) and the answer at this point is that we just donāt know for sure. CPU-Z is reporting a voltage of 1.482v from the Crosshair VI Hero motherboard when running at 4.0 GHz but I have seen that vary downward to as low as 1.417v, a significant drop if accurate.
i think this is interesting:
In addition to Lisa's comments, there are also some variables that could affect performance:
Ryzen benefits from disabling High Precision Event Timers (HPET). The timer resolution of HPET can cause an observer effect that can subtract performance. This is a BIOS option, or a function that can be disabled from the Windows command shell.
The bug with HPET has been known all along, AMD sent this info to reviewers (mentions of it in reviews). Several Intel platforms can have similar problem, my buggy X79 sees a measurable drop in perf if I turn on HPET in Windows. The problem is that AMD wants people to use their Ryzen Master overclocking tool, it throws up a warning dialog if HPET is off when you start Ryzen Master and tells you to enable it. HPET is off in windows unless you manually enable it.
The difference isn't big either, people have tested it and seen about a couple of percent or less of a difference. AMD has be quoted to say it can be a 5-8% difference, I've haven't seen any data to support this.
1 Like
So there were some reviewers that said you need a good CPU to let the new generation of GPU's stretch their leggs, right?
3 Likes
The benchmarks are looking good.
Then there are the comments, beyond cringe.
Yeah, the typical cancer section.
The Ryzen 1800X helped maintain the highest minimum frame rates weāve ever seen, and that means a more consistent, smoother and overall better gameplay experience. When it comes down to it, this higher minimum number is what you want from a gaming chip, not just the bigger average or maximum number.
This is my takeaway from many reviews and videos so far. Whatever you throw at ryzen, it will run smoothly. I hope that my memory kit arrives tomorrow, then I can try it out myself.
1 Like
The thing about ryzen is that people are testing it at stock clocks. When it is tested overclocked to 4ghz or higher it performs nearly as good as a 7700k. Then you have the optimizations coming down the line. My 1700 clocked at 3.8 no voltage increase runs smoothly in all my games. Rainbow Six Siege, CSGO, Paladins and Overwatch, Doom, Verdun.
That review gives very little useful data as they only use average FPS (practically worthless) and Minimum FPS (also mostly worthless). Frame Times is the interesting thing, you can calculate FPS from frame-time data if you want, but you can't get much usefulness from average FPS.
Didn't you link the TechReport Article earlier?
Look at their histograms. Compare a FX-8370 to Ryzen. Se how far beyond Ryzen is compared to Bulldozer/Piledriver of old. An FX-8370 might be able to get acceptable average FPS, frame-time data tell a different story.
Quote the article:
At the end of the day, I don't think it's worth putting too much stock in minimum frame rates. Our histogram analysis lets us see that they're extreme outliers that might not contribute more than a couple frames (if that) to the overall picture. We already have much better tools to make conclusions about component performance in the 99th-percentile frame time, frame-times-by-percentile, and plots of frame-time data that we present. We might start including frame-rate or frame-time histograms in our future reviews, however, because hey, they're interesting. Let me know what you think.
Oh, and yeah. Average FPS continues to be terrible. You can take that to the bank.
It doesn't even take much more work doing frame-time analysis today as we have tools like PresentMon. Sure, hardware capture like FCAT is better, but also much, much more expensive.
Edit: Ah yes, rant on and forget to write something that resembles a point on topic. Doh. I'm kinda eagerly waiting for frame time analysis with the Ti. TechReport has a beginning of an article, it is not complete yet. PCPer only tested with an Intel system, as they have for the last... too many years to think about. Hopefully there will be more.