*Official* NBR Desktop Overclocker's Lounge [laptop owners welcome, too]

I put it here also as it may effect some bench scores potentially in the future (plus, the people here can, after a fix is released, let us know the real performance hit). But, might be interesting on how AMD stacks up on some other performance metrics, depending on use (such as VMs) after the fix!

 

That's possibly bad news. I guess we will have to wait and see what the fallout actually looks like. Thankfully, I do not allow Windows Updates so I may not be affected by this hocus-pocus security baloney. My experience has been that Windows Updates can sometimes cause more harm than good. I am not convinced all of the blame rests on Intel. I think the Redmond Retards share at fair amount of the blame for their continued inappropriate influence and manipulation of what happens with hardware.

I won't be applying their "fix" - at least not until I am satisfied that it will not degrade my performance. That's more important than security as far as I am concerned.

And, how much of a winner would it make AMD if the primary way they get can finally get ahead in the performance race is based on an artificial metric driven by a frantic craze and childish clamor over a perceived need for a greater sense of security that imposes an artificial hit on performance? That's not a good way to get back on your feet and build a legacy based on excellent engineering. Everyone paying attention would know that it came about by the release of a screwed up Windows security patch.

 

Hmm, interesting. It affects Linux also, which leads me to believe that it is more on Intels shoulders than it is microshit.

"All that can be said at this time is that if this bug is as big as it sounds, it will be catastrophic for Intel"

 

Not enough known at this time, but we do already know that Micro$lop is a thorn in the side of Linux as well. Their Satanic push for UEFI Class 3 security filth is a ball and chain to Linux as well as Windows fanboys. Since they are an insignificant cog in the wheel of tech world, they get tossed to and from by the same tidal wave of sewage from the Redmond Mafia. Linux has had to bow to the Bastard King like everyone else. Micro$loth is the modern pipeline to trash tech.

 

Actually, it is all on Intel. Their predictive branch pre-fetch is the issue!

" How can this security hole be abused?
At best, the vulnerability could be leveraged by malware and hackers to more easily exploit other security bugs.

At worst, the hole could be abused by programs and logged-in users to read the contents of the kernel's memory. Suffice to say, this is not great. The kernel's memory space is hidden from user processes and programs because it may contain all sorts of secrets, such as passwords, login keys, files cached from disk, and so on. Imagine a piece of JavaScript running in a browser, or malicious software running on a shared public cloud server, able to sniff sensitive kernel-protected data.

Specifically, in terms of the best-case scenario, it is possible the bug could be abused to defeat KASLR: kernel address space layout randomization. This is a defense mechanism used by various operating systems to place components of the kernel in randomized locations in virtual memory. This mechanism can thwart attempts to abuse other bugs within the kernel: typically, exploit code – particularly return-oriented programming exploits – relies on reusing computer instructions in known locations in memory.

If you randomize the placing of the kernel's code in memory, exploits can't find the internal gadgets they need to fully compromise a system. The processor flaw could be potentially exploited to figure out where in memory the kernel has positioned its data and code, hence the flurry of software patching.

However, it may be that the vulnerability in Intel's chips is worse than the above mitigation bypass. In an email to the Linux kernel mailing list over Christmas, AMD said it is not affected. The wording of that message, though, rather gives the game away as to what the underlying cockup is:

AMD processors are not subject to the types of attacks that the kernel page table isolation feature protects against. The AMD microarchitecture does not allow memory references, including speculative references, that access higher privileged data when running in a lesser privileged mode when that access would result in a page fault.

A key word here is "speculative." Modern processors, like Intel's, perform speculative execution. In order to keep their internal pipelines primed with instructions to perform, the CPU cores try their best to guess what code is going to be run next, fetch it, and execute it.

It appears, from what AMD software engineer Tom Lendacky was suggesting above, that Intel's CPUs speculatively execute code potentially without performing security checks. It seems it may be possible to craft software in such a way that the processor starts executing an instruction that would normally be blocked – such as reading kernel memory from user mode – and completes that instruction before the privilege level check occurs.

That would allow ring-3-level user code to read ring-0-level kernel data. And that is not good.

The specifics of the vulnerability have yet to be confirmed, but consider this: the changes to Linux and Windows are significant and are being pushed out at high speed. That suggests it's more serious than a KASLR bypass.

Also, the updates to separate kernel and user address spaces on Linux are based on a set of fixes dubbed the KAISER patches, which were created by eggheads at Graz University of Technology in Austria. These boffins discovered [ PDF] it was possible to defeat KASLR by extracting memory layout information from the kernel in a side-channel attack on the CPU's virtual memory system. The team proposed splitting kernel and user spaces to prevent this information leak. Their work was reviewed by Anders Fogh, who wrote this interesting blog post in July.

That article described his attempts to read kernel memory from user mode by abusing speculative execution. Although Fogh was unable to come up with any working proof-of-concept code, he noted:

My results demonstrate that speculative execution does indeed continue despite violations of the isolation between kernel mode and user mode.

It appears the KAISER work is related to Fogh's research, and as well as developing a practical means to break KASLR by abusing virtual memory layouts, the team may have proved Fogh right – that speculative execution on Intel x86 chips can be exploited to access kernel memory."

 

So, why are we not hearing any clamor from the crApple fanboys about their Intel-powered turdbooks being at risk? Or, maybe they are having a hissy-fit and am I missing that they are simply because Mac OS X is such an unimportant element in the world that runs on Windows?

 

Actually articles are out saying Apple needs it too!

 

@Mr. Fox see the thing is, it IS bad news.

Even though you don't use updates or anything, the simple fact of the matter is, Windows as a "service" is killing the ability to use older versions easily. Hardware Acceleration was killed off for Youtube's VP9 codec in Windows 10 versions prior to Creator's Update, for example. It was an intentional choice. And more things are going to follow this route.

A lot of drivers and whatnot are going to be designed for newer versions of Windows. A lot of things are going to push you to update.

It doesn't much matter to YOU whose primary purpose of the machine is to benchmark, but it matters to a lot more people who have varied usage scenarios. People with 4K screens who consume a lot of 4K content for example MUST update. The CPU isn't enough. There's a lot of things that will force people to use newer OSes, and at some point, the OS will contain the change by default. Whatever comes after W10 FCU is certainly going to have the kernel change. And it'll eventually be required for smooth daily operation for a large number of people.

 

Why do you think I am slowly pulling away from this as a hobby. It's all turning to worthless crap, as you so eloquently explained. The more it becomes crap not worth doing, the less I am going to do it and the less money I am going to be interested in spending on crippled garbage. It's not about embracing change as much as it is about rejecting filth.

 

....and I was enjoying playing with my new toys too! Switching to Linux may be the future to staying away from the filth a little while longer?

 

https://www.fool.com/investing/2017/12/19/intels-ceo-just-sold-a-lot-of-stock.aspx

Don't worry guys I'm sure the Intel CEO sold all of his shares except those required to maintain CEO status just last month because he needed the money...

On a side note, it's already been reported that there is ZERO impact on gaming performance. Whew I can relax.

 

It's a bit early to relax, considering it's one test group sample on 8700K Coffee Lake, on Linux, with specific games:
http://forum.notebookreview.com/thr...-up-to-30-percent.812424/page-3#post-10657410
It's probably gonna be a little while (days?) before the details come out from embargo, which could help solidify your calm

 

Here's my AIDA64, I don't think I've seen anyone in these forums with a lower latency (although haven't looked hard!) (apart from Phoenix's old 6700K build), although I know latency isn't everything, but some say it's more important, this is only at 3200Mhz:

Congrats, it's a nice case, but can benefit from the mesh mod that GamersNexus did on it. Don't install the GPU vertically, GamersNexus showed that GPU temperatures rise significantly when vertically mounting GPU - that's more for watercooled cards that don't get negatively affected by a vertical mount.

 

Yeah, that is low latency. The CL14 and 1T definitely help with that. I see you are using the trial version. It would be interesting to know if the trial version only blocks the display of results or if part of the tests are incomplete. If the latter is true, I do not know if that might affect the reading on latency.

 

Yeah, 1T & CL14 help with that latency, I think a really low tRFC helps too, mine's only at 240 and 'normal' values are more around the 500 mark. I don't know whether trial version makes a difference to latency test, it's a little poorly designed if that would be the case, so I kind of doubt that - I'm not gonna buy AIDA, but maybe owners who own AIDA can temporarily install a trial version of one of the AIDA versions to do comparative tests to determine.

EDIT: during the testing, there is a pause at each stage where it says 'Trial Version' for the non-displayed tests - so it feels that the tests are happening, just the results not being displayed.

 

It's probably pretty close. Here is an example for comparison. I will try setting 240 for tRFC and see if it closes the gap on the 2.5ms difference in memory latency, but that is probably close enough to be within a margin of error as all of the memory readings will vary a little bit from one run to the next as well.

 

Cool, will be interesting to see your 240 tRFC. My RAM is not 'special', as it's not Samsung B-die, so people should be able to achieve my timings without too much fuss. In terms of 2.5ns being within repeatable error, I'd say that 0.2ns is more within repeatable error based on differences I've seen when running the test back to back.

 

My BIOS error checking mechanism will not allow me to use 240 for tRFC. It will halt and prompt me to change the settings because instability was detected. I set it back to the same and ran the test again and the latency gap closed by 0.9ms with identical settings. Mine is not Samsung B-die either. It's Hynix. But, it could be due to the CPU being clocked at 5.2GHz as well. Latency does seem to improve with slightly lower core clocks, while read/write/copy speeds typically improve slightly.

 

Interesting you saw latency fluctuate that much, on my system it's normally within 0.2ns. Another reason for the low latency seen on mine could be the difference between 8700K 6 core vs 6700K 4 core - perhaps a larger ring bus increases the latency. Also, I have my cache at 4.7 Ghz too, and people don't normally overclock their cache - so my CPU & cache are both at 4.7Ghz. Maybe motherboards & motherboard firmware differences affect latency too, as well as how many RAM sticks (I have 2 sticks).

EDIT: although, actually now I think about it & remember my testing, the low tRFC is the main reason for that extra lower latency, I had higher latency with 'normal' high tRFC's.

 

Yes, I think all of those things can have some effect on latency. 7700K and 8700K do not let the ring run equal to core like 6700K. You can set them equal and the ring will always run 300MHz less by default. With 6700K the ring and core voltage could also be set independently, and that ability was lost with 7700K and 8700K. Their voltage settings are synced to match unless you use ThrottleStop to set them to manually override them with different values.

 

This doesn't work for me. I can't read cache voltage, but I AM certain every time I set cache higher than core, core rises. It's going to take the higher of the two values as far as all my testing has produced.

 

Yes, that is exactly what happens. Cache voltage will override core voltage if you set it higher. But, there is no reason to set cache voltage higher than core.

 

So are you certain cache voltage will drop if it's set lower? Because I can't or don't know how to test

Sent from my OnePlus 1 using a coconut

 

Why is my memory latency so high? Is it because I am on Ryzen?

 

Because Ryzen has the latency between cores, but the added latency of CCX comms. It is the nature of the beast on that design for the processor.

 

Overclock the core higher and you will see it. If you set cache voltage too low it will cause Windows to crash because it will pull down the core with it. @Johnksss demonstrated that a long time ago, but I don't remember if it was in this thread.

 

Hmm... I see. Interesting point, good to know. I don't think I'll tell too many people though. That's the kind of thing that can go wrong real fast for a user who doesn't know what they're doing

 

I bought this case, because it has the ability to add a 360mm radiator to the top, or the front of the case. The solid piece of tempered glass for the side panel, is extremely nice! Its a bit on the heavy side, but it looks very sharp. I know the temps at Idle and overclocked have not changed enough to notice.

 

Temps haven't changed in comparison to your previous case? What case did you have before?

 

I had a CoolerMaster Master box. Idles at 30c max at 60c, Overclocked maxes out at 71c

Update: the new Agesa/Bios update I did sure bounced my cinebench scores up!

 

That's not bad I guess with air cooling. My CPU is at 20 degC as I type this, which is equal to the room temperature, games in the 30 to high 40's depending on game (average temperatures, not peak), and benches in the low 60's - but I do have that beast of an air cooler & a liquid metal delid, and a CPU which can only really chew up to 130W in Prime95, as opposed to your 8 core monster - I think my case is good for CPU temperatures though too.

 

Yes, thats why I purchased this case to put a water cooled loop in, and I have always liked cooler master products.

 

What bios/agesa versions?

 

Im sure its not new for you, as I hadnt got around to updating my board yet. It is the one released in September, so I am sure you have it already.

 

Actually, no. Agesa is different between the AM4 and TR4. Although from September the updates of AM4 may have made it too TR4.

 

Gaming And Streaming: Which CPU Is Best For Both? - Tomshardware.com
"Many streamers place video quality over maximizing the frame rate of whatever game they're playing, so your own priorities will largely dictate how you tune your system. In fact, turning on v-sync may be a good way to balance streaming and gaming performance.If you seek the highest in-game performance while you stream, Intel's Coffee Lake-based Core i7-8700K is a good fit. The Ryzen 7 1800X is also competitive and tends to offer better streaming performance. Using our settings, the 1800X also had more CPU headroom leftover for more taxing encode settings, if desired. Granted, some of that extra horsepower is due to the 1800X's lower gaming performance, which means there are fewer frames to encode."

 

AIDA64 Memory Test @ 3466

 

Nice scores. What clock speed? Are you supplementing your liquid cooling with air conditioning? Does it seem that unparking the cores helped a little bit?

 

Yup, you did it right. I replied to your email. Unparking the cores will have nothing to do with 3DMark 11 or Vantage crashing. All that does it blocks Micro$loth cancer from trying to suppress CPU performance to reduce power consumption. You've got something else going on. If Cinebench is running that good, its not the CPU. At what point in the Vantage or 3DMark 11 run is the crash happening, and what behavior are we referring to as a crash? BSOD or something else? More likely than not, it's either a buggy NVIDIA driver or your GPU overclock is unstable/too high (or a combination of crappy unstable driver/high overclock).

 

I don't think Windows 10 parks cores does it? I checked in Windows Resource Monitor under CPU and it doesn't say they're parked. It used to say they were parked if I remember back to Win 7.

 

Yes, it absolutely does. Worse than Windows 7.

 

Run the GPUs stock and see if it still happens or if the benchmarks run normally.

 

Fair enough. Do you know where I can see if they are parked or not? It used to show them in the Resource Monitor back in Win 7.

 

Always best to change only one thing at a time in the process of elimination so you can isolate what is causing the issue.

Vantage exited to the desktop even with everything stock? If so, switch over to the other OS and see if the same thing happens. If it only happens under Windows 10 or 7, and not both OSes, that information could be relevant.

 

Quick question. Since I am running Windows 7, and benchmarking say the passmark suite, will that hinder my score? I know there is a DX12 test on it. Does it just bypass the test for direct x 12? Or does its still run the test since my card is Dx12? Just a bit confused about that.

 

https://www.theguardian.com/science...red-with-more-than-23m-digits?linkId=46589739
P95 FTW! GIMPS, which makes prime95, just found the largest prime number ever discovered!

 

Use this to see them parked and unpark them. Link to my Google Drive...

https://drive.google.com/file/d/0Bwdqi25LDwZyVExIUV9HS045RDg/view?usp=drivesdk

 

I do not think so. As far as I know it will not. Since I am dual-booting both I can check and find out.

 

This guy has to be one of the biggest idiots on the interwebs yet. He's peddling this non sense to the uninformed sheep. People actually believe what he's saying, and if you listen much of what the guy says is either very deceptive or complete BS. Anyone else seen some of his videos? It's like he doesn't actually understand what TDP actually is.

 

Yeah, I have seen that guys videos many times. That clown has always been an AMD shill for years, including the horrible pre-Zen garbage and GCN trash GPUs, LOL. The sad part is I think he actually believes his own nonsense.