ASRock X299 Taichi

[eelmor]

A few more things that would make the review more informative. Most pages are 1-2 pictures, a few sentences and perhaps a table.

https://www.techpowerup.com/reviews/ASRock/X299_Taichi/ and https://www.techpowerup.com/reviews/ASRock/X299_Taichi/3.html

Memory: 8x DIMM, Max. 128 GB; supports 4400 MHz+(OC)

"ASRock claims crazily high 4400 MHz memory support for these eight DIMM slots. This board can hold a maximum of 128 GB."

Should be noted that this depends on if you plug a Kaby Lake-X or Skylake-X CPU. Highest memory QVL on Skylake-X is 4133 MHz, max capacity on Kaby Lake-X is 64GB.

https://www.techpowerup.com/reviews/ASRock/X299_Taichi/5.html

No testing of the software?

https://www.techpowerup.com/reviews/ASRock/X299_Taichi/6.html

Would be nice with some pictures of the fan control options and what you can do, like what kind of fan curves, which temperatures you can map the fan speed to and range for the temperatures. Sometimes, I know for example on our boards, the temperature range can be limited in some cases. I assume the fan speed control ranges in the table are in PWM mode, how about in DC mode?

https://www.techpowerup.com/reviews/ASRock/X299_Taichi/8.html

As my previous post, more details please. Where/how were the measurements taken? What does your test setup look like? Airflow? Ambient temperature?

https://www.techpowerup.com/reviews/ASRock/X299_Taichi/9.html

"ASRock's last generation of Taichi boards might have had some audio issues when testing. I had talked to ASRock about this kind of thing in the past and got the response, "I guess we'll just have to do better". Bill and Ted would be proud. You young folk wouldn't understand."
Same, more details please.

https://www.techpowerup.com/reviews/ASRock/X299_Taichi/10.html

You mention that there are several controllers on the board, but not which ones are used in the tests?

https://www.techpowerup.com/reviews/ASRock/X299_Taichi/11.html

"Overclocking on the ASRock X299 Taichi is going to be quite largely influenced by the cooling you employ. High clocks are possible to all cores, but if you use a SkyLake-X CPU, be prepared for some extreme power usage once you crank up the clocks. Memory clocking is effortless with these CPUs and this board, too."

Where's the testing and data to support all this?

https://www.techpowerup.com/reviews/ASRock/X299_Taichi/12.html and https://www.techpowerup.com/reviews/ASRock/X299_Taichi/13.html and https://www.techpowerup.com/reviews/ASRock/X299_Taichi/14.html

What's the theory behind these performance benchmarks? Why were they chosen? What are they supposed to tell the customer? Why are there no other X299 boards for comparison?

https://www.techpowerup.com/reviews/ASRock/X299_Taichi/15.html

I've tested the X299 Professional Gaming i9 which is essentially the same, and I agree that it's one of the better boards available at launch compared to certain other brands. I'm curious to know if you've tested any other motherboards? I'd like to see something in the review regarding how they compare to this, since you're giving it the Editor's choice award.

 

[[XC] Oj101]

Another question:

Memory clocking is effortless with these CPUs and this board, too.

All I see, however, is DDR4-3600 CL16 running at, well, DDR4-3600 CL16. What sort of memory overclock did you get?

 

[cadaveca]

Will you update with performance comparisons against boards that are actually on the same platform?

AS time progresses, probably. I make no guarantees though.

How was temperature and power measured during the VRM test? Can you post a picture of the test setup?

Thanks

Fluke clamp-on meter over EPS lines and IR thermometer. And for the billionth time, no, I am not posting pictures of this again. You can find them on here somewhere. These two tools cost less than $150 and are hardly a big deal and should be in every reviewer's arsenal. To be honest, I'm shocked you are even asking. I think you pointed me in the right direction for the clamp on meter a decade ago. BUt now we don't agree on something, and this is the angle you take? OK.

cadaveca has some odd ideas for pros and cons. Supporting tri-SLI is apparently a bad thing, as was having LN2 features on the Maximus IX Formula. I'm not sure how he comes up with them xD

That said, cadaveca, if the review was done on a prerelease BIOS surely one of two things should happen:

1. You should knock off points for every flaw on the prerelease BIOS with a statement at the end that a new BIOS may fix things and the review is in no way representative of the final product
2. You should wait for the BIOS before doing the review so the review actually has relevance

Don't you think?

1. one flaw; M.2 performance. Everything else worked as it should.
2. board maker requested quick review, so it was done... on the 21st of June. Not my fault it wasn't posted till now. Had it been then, before the boards went on sale on the 26th, it would have highly been relevant, but...


And yeah., my cons may seem weird... gives some thing to think about. No single board is going to be right for everyone, anyway, simply by budget constraints...

How's the Uncore overclocking with this board? I heard it gives a nice boost to performance.

That depends on each CPU, so far that I have seen. Not enough chips to really test to see any differences

I've tested the X299 Professional Gaming i9 which is essentially the same, and I agree that it's one of the better boards available at launch compared to certain other brands. I'm curious to know if you've tested any other motherboards? I'd like to see something in the review regarding how they compare to this, since you're giving it the Editor's choice award.

You're first comments, most of those questions are answered by the provided tables. Actually, I see you read the review, but it seems you didn 't, because many of your questions here are actually answered in the review. Overclocking? what you see is what you get.

Software? I test Out-of-Box experience In this, there is no software other than the tools shown. Controllers? listed in the table above.

I see where you are coming from, but I'm not writing my reviews for YOU. You already know all this, and don 't really need reviews. It's very weird to me, honestly, to have a guy that benches in elite league on hwbot is here asking me about reviews.... lol, one of the guys pictured in ASUS RealBench... a guy that posted 7900X CPU on HWbot back in May?

https://hwbot.org/submission/3564060_elmor_xtu_core_i9_7900x_4059_marks


Perhaps you're trying to save face for ASUS? Like, I gotta call ya out on this one.

 

[eelmor]

Fluke clamp-on meter over EPS lines and IR thermometer. And for the billionth time, no, I am not posting pictures of this again. You can find them on here somewhere. These two tools cost less than $150 and are hardly a big deal and should be in every reviewer's arsenal. To be honest, I'm shocked you are even asking. I think you pointed me in the right direction for the clamp on meter a decade ago. BUt now we don't agree on something, and this is the angle you take? OK.

Your numbers seem a bit different than mine, so I want to try to figure out why. Any results in a review should come with enough details for replication. You're not providing enough information to do this. I asked for a picture of the setup mainly to be able to get information on the airflow situation. What's your ambient temperature? Where are you measuring the VRM temperature with your IR thermometer? 253W input to the VRM is approximately 220-230W actual CPU power consumption, will retest using this scenario next week.

 

[cadaveca]

OK, this I can deal with. ambient is corrected to 18c. Actual ambient during this test was 18.6c.

I measure the entire area, and report the highest temp I can find, same for PCH cooler. SO I measure chokes, edge of fets if they can be seen, cooler, and PCB real estate between components.


load testing is wprime1024M; also used in our cooler reviews for a bit of correlation between reviews.

 

[DRDNA]

outa the wood work the flocks stop in for a bite.....

 

[Steevo]

cadaveca has some odd ideas for pros and cons. Supporting tri-SLI is apparently a bad thing, as was having LN2 features on the Maximus IX Formula. I'm not sure how he comes up with them xD

That said, cadaveca, if the review was done on a prerelease BIOS surely one of two things should happen:

1. You should knock off points for every flaw on the prerelease BIOS with a statement at the end that a new BIOS may fix things and the review is in no way representative of the final product
2. You should wait for the BIOS before doing the review so the review actually has relevance

Don't you think?

So many people reviewed Ryzen and well.... https://www.techpowerup.com/reviews/AMD/Ryzen_7_1800X/16.html I know this is for motherboards, but if we are being fair, Intel should be knocked for "prerelease" anything, poor drive performance, lack of integrated graphics, and anything else, like the strange win loss from older chips with more cache.

 

[MrGenius]

As to voltage, that's Input voltage; these CPUs are like Haswell, and voltage control is inside CPU.

Hmmm...it seems like I've seen CPU-Z show it differently in one of der8auer's vids. If you watch the vid between 2:12 and 2:34 he shows his delidded 7900X running P95 @ 5.0GHz on 10 cores 100% load with CPU-Z v1.79.0 showing 1.320V.

But then in another one of der8auer's vids it, at a couple points, shows something more like what you've shown. If you watch that vid between 6:36 and 6:46 it shows a 7900X also running P95 @ 4.4GHz on 10 cores ???% load(up to 267W) with CPU-Z v1.79.0 showing 1.728V(actually fluctuating between 1.704V and 1.740V). Then between 7:05 and 7:11 it shows the same 7900X running P95 @ 4.4GHz on 10 cores ???% load with the VCCIN increased by .2V and CPU-Z v1.79.0 showing 1.860V(actually fluctuating between 1.836V and 1.896V). However, at another point, between 18:07 and 18:20 again a 7900X is shown running P95 on ?? cores ???% load throttling between 4.6GHz and 1.2GHz with CPU-Z v1.79.0 showing 1.188V.

So I'm totally fricken lost. I'll admit I know nothing about Haswell or Skylake-X CPU core voltage control(never owned either one). But I thought I understood the concept of Vcore in general. I guess I don't.

 

[cadaveca]

So I'm totally fricken lost. I'll admit I know nothing about Haswell or Skylake-X CPU core voltage control(never owned either one). But I thought I understood the concept of Vcore in general. I guess I don't.

Voltage is supplied to different parts of the chip at different levels. In some platforms, the board regulates this, but in some platforms, the voltage regulation is inside the CPU, and the CPU receives a single voltage from the board to "downstep" all those different voltages to. VCore is still VCore, but whether it is applied to the CPU by the board or the CPU itself is the difference. With voltage regulation inside the CPU, voltage control can be more precise since it travels down less "wire", but the offset of that is that the CPU die itself runs hotter. Hence the hole "OMG THIS CHIP RUNS HOT" stuff.

CPU-Z allows you to change the sensor it displays, and on different boards, it shows different sensors. This is one of those cases of "never trust software".

 

[DRDNA]

Voltage is supplied to different parts of the chip at different levels. In some platforms, the board regulates this, but in some platforms, the voltage regulation is inside the CPU, and the CPU receives a single voltage from the board to "downstep" all those different voltages to. VCore is still VCore, but whether it is applied to the CPU by the board or the CPU itself is the difference. With voltage regulation inside the CPU, voltage control can be more precise since it travels down less "wire", but the offset of that is that the CPU die itself runs hotter. Hence the hole "OMG THIS CHIP RUNS HOT" stuff.

CPU-Z allows you to change the sensor it displays, and on different boards, it shows different sensors. This is one of those cases of "never trust software".

Well stated

 

[EarthDog]

Voltage is supplied to different parts of the chip at different levels. In some platforms, the board regulates this, but in some platforms, the voltage regulation is inside the CPU, and the CPU receives a single voltage from the board to "downstep" all those different voltages to. VCore is still VCore, but whether it is applied to the CPU by the board or the CPU itself is the difference. With voltage regulation inside the CPU, voltage control can be more precise since it travels down less "wire", but the offset of that is that the CPU die itself runs hotter. Hence the hole "OMG THIS CHIP RUNS HOT" stuff.

CPU-Z allows you to change the sensor it displays, and on different boards, it shows different sensors. This is one of those cases of "never trust software".

you can trust that its the input voltage!!!

Genius... FIVR - https://www.google.com/search?q=wha...droid-verizon&sourceid=chrome-mobile&ie=UTF-8

 

[efikkan]

It looks like a pretty decent board, but it lacks 10 Gbit/s Ethernet which would make it a great workstation board. Still, it includes WLAN, which is a complete waste for such a board.

 

[RealNeil]

Good review, Dave.

 

[Hossein Almet]

It's now official, all those reviewers who have said that the 7900X can clock up to 4.8GHz with a voltage of only 1.3 V were lying, or didn't know what they were talking about.

 

[cadaveca]

It's now official, all those reviewers who have said that the 7900X can clock up to 4.8GHz with a voltage of only 1.3 V were lying, or didn't know what they were talking about.

Well, honestly, they probably could get close, but the heat kinda kills that from being an option.

 

[RealNeil]

Well, honestly, they probably could get close, but the heat kinda kills that from being an option.

Unless you're nostalgic and want an easy-bake oven like your sister had as a kid.

 

[dwade]

I believe the Taichi x299's 8 pin is throttling the 7900x. Many people are saying it needs 8+4 pin to work well with that 10 core nuclear warhead. Asrock have already released a BIOS that fixed the turbo 3.0 issue so we shouldn't see regressed performance versus 6950k.

I mean in this review:
https://www.overclock3d.net/reviews/cpu_mainboard/intel_core_i7_7820x_skylake_x_review/17

The Taichi was able to tame the 7820x pretty well and we saw it slaughtering the entire Haswell-E lineup due to no throttling. There were no signs of huge performance penalty compared to the findings from TPU's review.

I believe no BIOS will fix this issue when using 7900x with any x299 boards with a single 8 pin. 7820x even when overclocked to 4.7Ghz should have no problems however.

 

[cadaveca]

I believe the Taichi x299's 8 pin is throttling the 7900x. Many people are saying it needs 8+4 pin to work well with that 10 core nuclear warhead. Asrock have already released a BIOS that fixed the turbo 3.0 issue so we shouldn't see regressed performance versus 6950k.

I mean in this review:
https://www.overclock3d.net/reviews/cpu_mainboard/intel_core_i7_7820x_skylake_x_review/17

The Taichi was able to tame the 7820x pretty well and we saw it slaughtering the entire Haswell-E lineup due to no throttling. There were no signs of huge performance penalty compared to the findings from TPU's review.

I believe no BIOS will fix this issue when using 7900x with any x299 boards with a single 8 pin. 7820x even when overclocked to 4.7Ghz should have no problems however.

You're funny. There's no throttle if you cool the system properly. I finished re-benching yesterday with most recent BIOS, both stock and OC clocks, no real issues to report.

you don't NEED 8+4, but yes, some CPUs might exceed the power that an 8-pin plug can supply, but for most CPUs, they will not be able to do so unless you de-lid, IMHO. The platform has a base spec of 300W.

 

[EarthDog]

Did you read the toms article dave? I think its very well done...

http://www.tomshardware.com/reviews/-intel-skylake-x-overclocking-thermal-issues,5117.html

One takeaway... you are thermally limited by the cpu before the vrm in the vast majority of cases..

 

[cadaveca]

Feedback is good, but let me give you some feedback on your feedback, so you understand why I have made some adjustments based on your feedback, but not many.

@W1zzard

A few more things that would make the review more informative. Most pages are 1-2 pictures, a few sentences and perhaps a table.

No, about 1/3 of the pages do. You are exaggerating. I do put a minimal amount of words in a review so that there is no personal opinion.

We do also look at which pages people visit the most, and I have put more focus there.

Should be noted that this depends on if you plug a Kaby Lake-X or Skylake-X CPU. Highest memory QVL on Skylake-X is 4133 MHz, max capacity on Kaby Lake-X is 64GB.

That's fair for sure.

No testing of the software?

What software? the app that downloads software? The reboot function? That's all there was. You can see form the screenshot of the App Shop that none was available at the time of testing. I have, of course, tested on other boards, so know what to expect. Such is the problems of pre-release testing.

Would be nice with some pictures of the fan control options and what you can do, like what kind of fan curves, which temperatures you can map the fan speed to and range for the temperatures. Sometimes, I know for example on our boards, the temperature range can be limited in some cases. I assume the fan speed control ranges in the table are in PWM mode, how about in DC mode?

Sure. Some of this is shown in the BIOS page, since that's a BIOS function. But you're right, it could be duplicated here as in other reviews.

As my previous post, more details please. Where/how were the measurements taken? What does your test setup look like? Airflow? Ambient temperature?

Test set-up is shown in pictures? Ambient and airflow... ambient, yeah, could put a disclaimer about temperatures being normalized, but airflow? Am I testing a board or a cooler? I'm not sure what you mean here...?

"ASRock's last generation of Taichi boards might have had some audio issues when testing. I had talked to ASRock about this kind of thing in the past and got the response, "I guess we'll just have to do better". Bill and Ted would be proud. You young folk wouldn't understand."

Same, more details please.

Graphs say it all; Bill and Ted's Excellent Adventure (LINK) But yeah, some pop culture references won't be known by all users, so I understand why you don't get it.

You mention that there are several controllers on the board, but not which ones are used in the tests?

Uh, did you look at the table?

"Overclocking on the ASRock X299 Taichi is going to be quite largely influenced by the cooling you employ. High clocks are possible to all cores, but if you use a SkyLake-X CPU, be prepared for some extreme power usage once you crank up the clocks. Memory clocking is effortless with these CPUs and this board, too."

Where's the testing and data to support all this?

Um, testing and data? We're talking about how easy it was for this board to OC. You change BIOS options, and they work, or they don't? You enable XMP, and it works, or it doesn't? I'm not benching here for records....

What's the theory behind these performance benchmarks? Why were they chosen? What are they supposed to tell the customer?

OK, this, I don't understand... you're in elite leagure at HWBot, and are asking what do benches actually show?

Why are there no other X299 boards for comparison?

There never is on my first published review of a platform, going back many years now. I'm not comparing this board to other X299 boards, I'm comparing it to all boards ever in existence.

I've tested the X299 Professional Gaming i9 which is essentially the same, and I agree that it's one of the better boards available at launch compared to certain other brands. I'm curious to know if you've tested any other motherboards? I'd like to see something in the review regarding how they compare to this, since you're giving it the Editor's choice award.

Yeah, I got MSI and more ASRock boards. I do not compare boards head-to-head, because a $100 board isn't meant to compete wit ha $400 board. I am considering this board and ASRock's intended audience, and whether it meets those customer's needs.


See, I have to say, you are stepping out of the extreme clocking world, but not all reviews need to have a focus that leads towards performance metrics. If I had my way, there would be no benchmarks at all in my motherboard reviews.

That's not to say you don't have valid points, but what's valid to you are a PC user isn't going to be valid to ALL PC users, unfortunately. I try to cater to a different user than you, who works in the industry, gets hardware for free, and is into competitive benchmarking. You don't need my reviews.

But you still have some good feedback, so I will be making some adjustments based on your direct feedback, but I don't think they are the exact ones you want. [/quote]

 

[RealNeil]

Feedback is good, but let me give you some feedback on your feedback, so you understand why I have made some adjustments based on your feedback, but not many.

@W1zzard

But you still have some good feedback, so I will be making some adjustments based on your direct feedback, but I don't think they are the exact ones you want.

Tweaked review or not, it was still a good read in it's original form.
It really ~is~ possible to take a review at face value and not try to pick the fly shit out of the pepper.

Right?

 

[cadaveca]

Tweaked review or not, it was still a good read in it's original form.
It really ~is~ possible to take a review at face value and not try to pick the fly shit out of the pepper.

Right?

Oh, it's not this review that gets adjustments; it's just things I will make sure to have in future ones.

 

[[XC] Oj101]

Um, testing and data? We're talking about how easy it was for this board to OC. You change BIOS options, and they work, or they don't? You enable XMP, and it works, or it doesn't? I'm not benching here for records....

To be fair I don't see ANY memory clocking in your review, but you state that it's effortless. Enabling XMP is not overclocking, it's running the memory at its stock rated speed.

 

[cadaveca]

To be fair I don't see ANY memory clocking in your review, but you state that it's effortless. Enabling XMP is not overclocking, it's running the memory at its stock rated speed.

Yep, but clocking and overclocking aren't the same word, are they?



Also, I come from the world where Samsung makes 2400/2666 MHz ICs, and memory makers overclock them and sell them @ much higher speeds, so we just gonna go back and forth on this. What the memory OEMs do isn't very different from what I'm doing.

I don't cover OC in greater detail because each person's results will vary greatly based on what other stuff they install. Someone can buy the exact same hardware I use, and get far worse results. Yet a motherboard makes specific claims as to memory support, which I do test. Just because the results aren't shown doesn't mean I didn't do it... there's something to be said about the economy of words.

I could drop screenshots, but why? I like the discussion. Elmor can be critical about boards being boring, yet he also says they sell fine, so what exactly is the problem? A dude that works in the industry has gotten board(sic ) with his job, and we are all supposed to react? Please.

Those chosen settings I have for each platform on OC does show differences in board's BIOS and how they react to the exact same hardware and settings, and how that affects the end user, will be something that is mentioned. You'll see more in the next X299 review. I didn't just review this board and call it good...

Earthdog is waiting for more boards? Err... why? I know why...

 

[[XC] Oj101]

Yep, but clocking and overclocking aren't the same word, are they?



Also, I come from the world where Samsung makes 2400/2666 MHz ICs, and memory makers overclock them and sell them @ much higher speeds, so we just gonna go back and forth on this. What the memory OEMs do isn't very different from what I'm doing.

I don't cover OC in greater detail because each person's results will vary greatly based on what other stuff they install. Someone can buy the exact same hardware I use, and get far worse results. Yet a motherboard makes specific claims as to memory support, which I do test. Just because the results aren't shown doesn't mean I didn't do it... there's something to be said about the economy of words.

I could drop screenshots, but why? I like the discussion. Elmor can be critical about boards being boring, yet he also says they sell fine, so what exactly is the problem? A dude that works in the industry has gotten board(sic ) with his job, and we are all supposed to react? Please.

Those chosen settings I have for each platform on OC does show differences in board's BIOS and how they react to the exact same hardware and settings, and how that affects the end user, will be something that is mentioned. You'll see more in the next X299 review. I didn't just review this board and call it good...

Earthdog is waiting for more boards? Err... why? I know why...

In the 15 years I've been in the hobby this is the first time I've heard "clocking" not mean overclocking - it's never meant "running at stock" so forgive me for not understanding what you meant.

If we go with your definition as "running it out of the box" what's the significance of it? Show me a Z170/Z270/X299 board that can't run XMP, cause I don't know of one? If you were using a kit of, say, DDR4-4800, would you only run it at DDR4-4400 as that is the maximum the ASRock claims to support? And as per the specs it supports DDR4-4400 (OC), so why not test that? If you're not testing the speeds marked as "(OC)" then why test past DDR4-2666, as even DDR4-2800 is listed as "(OC)"? If you "don't test overclocking" and at the same time believe that G.Skill overclocks the Samsung B-die ICs, why are you testing anything above the Samsung rated speeds? This seems more like "this board was capable of running the frequency of a random set of memory I had lying around" than testing whether it can do as claimed, as your testing lies between two claims (non OC and OC).

Additionally, if you "don't test overclocking" because results vary from setup to setup, why do you test CPU overclocking at all? I come back to you saying that "Just because the results aren't shown doesn't mean I didn't do it... there's something to be said about the economy of words." yet I have to ask why show the DDR4-3600 if you did indeed test up to DDR4-4400?

We're talking about how easy it was for this board to OC. You change BIOS options, and they work, or they don't? You enable XMP, and it works, or it doesn't? I'm not benching here for records....

You can pretty much ignore that quote, I'm just pointing out that you've once again referred to testing overclocking.

I need to address several other points I saw now, so this post may be a little long. I'm addressing them in no particular order, either.

OK, this, I don't understand... you're in elite leagure at HWBot, and are asking what do benches actually show?

Just because elmor understands doesn't mean that everyone understands. Is it not good for someone who knows the right questions to ask said questions for those who don't? An average reader (I'm not patronizing the TPU membership, I'm talking truly average from whichever walk of life) may not know that the right questions even exist, so is it not a good idea for someone who knows the questions to ask them?

I'm also an overclocker, I'm also in elite league, so what? The review isn't aimed at us. We don't read a review and decide to buy a board. Reviews help those who are not in this position to choose a board, and chances are they don't have a fraction of the understanding to turn benchmark results into an informed decision. You're saying "these are the results, they're good" but not explaining the how or why.

There never is on my first published review of a platform, going back many years now. I'm not comparing this board to other X299 boards, I'm comparing it to all boards ever in existence.

Then is the better option not to hold back the review until you can make a worthy comparison? Sure, you lose the "appeal" of being first with the review, but is a worthwhile review not worth more?

Yeah, I got MSI and more ASRock boards. I do not compare boards head-to-head, because a $100 board isn't meant to compete wit ha $400 board. I am considering this board and ASRock's intended audience, and whether it meets those customer's needs.

That is more than fair enough, but it doesn't explain how you can give it an Editor's Choice award when, for all anyone knows, it ends up being the worst X299 board out there. You've given this board a 9.8 out of 10 - honestly, what would a board score if it were to offer all of the features of the Taichi plus more (maybe another four SATA ports, easier overclocking, another M.2 slot, four extra USB 3.0 ports, whatever the case may be)? Would it score above 10, would the score of this board drop (as well as the Editor's Choice award fall away), or would it really only be able to score a MAXIMUM of 0.2 more than this? By giving it an Editor's Choice award straight off the bat you've limited your future review scoring unless you come back and edit this review, in which case was it reliable to begin with (rhetorical question)?

If I had my way, there would be no benchmarks at all in my motherboard reviews.

Then I have to ask what would set them apart? "MOAR RGB!!!"? There are only so many metrics that can be used to judge a board - appearance, I/O and performance being the main three. Appearance is too subjective, and I/O is something that can be read on paper (in which case list the specs of every X299 board and give the award to the board with the most I/O connectivity options). That leaves performance, which you want to leave out? Performance is one of the main things setting one board apart from another!

I see where you are coming from, but I'm not writing my reviews for YOU. You already know all this, and don 't really need reviews. It's very weird to me, honestly, to have a guy that benches in elite league on hwbot is here asking me about reviews.... lol, one of the guys pictured in ASUS RealBench... a guy that posted 7900X CPU on HWbot back in May?

I don't see what your point is. It's no secret at all who elmor is or where he works. The people who buy the boards he helps develop are not him, and 99.9 % don't have the understanding he has. Is it not in the best interests of everyone to make sure that the reviews are accurate and helpful to the 99.9 % ?

Such is the problems of pre-release testing.

Pre-release testing has indeed got problems as we've already discussed - but pre-release is only handy to those who actually get the hardware pre-release. Once the release hardware is on the shelves and in Joe Soap's computer it isn't necessarily the same kettle of fish as the pre-release version, so is the review even applicable to Joe Soap? If you're aiming your review at Joe Soap you've got to make sure that your experience will as closely exactly match his as possible, which isn't possible with a pre-release review.

I think that's it for now.