Discussion in 'Article Discussion' started by bit-tech, 30 Sep 2019.
I'm not sure what to make of this review, comes across as biased/uneven.
A few things I've noted:
1) Board does come with ssd thermal pads.
2) overclocking section at the start of the paragraph you state that the board doesn't run at 4.3ghz overclock like the rest of the boards but then saying actuary it does you just need to adjust load line calibration. Why the negative spin on this. In addition this is the only board that's itx in the tests isn't it so will have limitations the others don't, would have been better to start with it reaches the same clocks but additional config is needed. LlC in no way is a bad thing for normal use so dont believe this is a con.
3) Stating bad vrm and ssd temps without actually posting numbers and then saying the crosshair viii formula impact is better when this is a bad comparison. It's a mini dtx board with more room for cooling and they have a different location for the X570 chipset on that board.
All in all I think this ubfortunetly is a poor review which is a shame as bittech is/was a go to site for me.
I didn't write the review - and I haven't read it, either, not being in the market for a humpty-tump-hundred-quid motherboard - but I can tell you that when I am reviewing things I review 'em as-provided. If you can twiddle things to make it better, I'll say you can twiddle things and tell you what difference it makes - but I'll tell you how it was out-the-box first. That'd go doubly if I had a bunch of near-identical things on the testbench and you only needed to twiddle a specific thing on one of 'em - and I'd be asking why only that one needs the twiddle twiddling.
I'd say "out of the box" is important, but at the same time, I think the audience here wouldn't normally mind a bit of tweaking - hell, a lot of us, always seem to be tweaking settings, so it definitely isn't a negative. Don't think the fact it's ITX should allow negatives to be glossed over though.
After reading the review, I don't think it comes across a biased or uneven, on the issues you raised, thermal pads for m2 are mentioned as included, and it's rightly pointed out that there are issues with them. Overclocking and using LLC, and saying, they've had to do it with Gigabyte boards before - useful information for anyone unsure of where to start looking to get better speeds, or wondering why everyone else *is* getting better speeds . Would be good to get measured temps when mentioning if they seem particularly high or low though, agree with you there.
In relation to using stock setting/auto that's fine for out of the box but your going outside of the warranty and AMD specification when overclocking. So one board needing x number of tweaks versus another needing y number has very little to do with it if both reach the same overclock and at same voltage. Its just one goes about it slightly differently. Now if the overclocking was difficult on the Gigabyte board and/or the bios options weren't there to provide the means then this would be an issue. And as you state youd ask why the twiddle needs twiddling but in this article it makes no mention only that its a negative... when in fact LLC is quite a positive thing for everyday usage to prevent overvoltage during switching from high load to low load scenarios and vice versa.
Stating you need to put LLC to extreme is informative, but I think the piece should have been written with a positive introduction that the board meets the same overclock as other x570 boards however please note that LLC needed to be tweaked. Why start with a negative spin that it wasn't stable? Maybe I just took the tone the wrong way easily done online.
On the thermal pads page 1 says "is that there are no thermal pads for the M.2 SSD and thus no proper thermal", but they are included and Ive used the board and haven't experienced temperature issues with an adata xpg8200 pro ssd.
Temperatures and the general comparison to other X570 boards without highlighting that they are ATX or larger format mini-itx baords is also not brilliant reviewing. This has quite an impact on overall results due to space limitations/vrms used etc.
I've tested alot of x570 boards and there are pros/cons to them, but alot of the negatives in this review are compared to a 400 sterling mini DTX board that isnt the same form factor never mind the same price bracket.
Maybe its just me but ive read more well rounded reviews form other sites and its a shame as I generally like bit-tech
Firstly I would point out that board got an award. We don't hand those out to products that are poor as many of these comments are suggesting. However I feel I need to address some of the concerns raised here as some deserve answered or are simply inaccurate.
1) It comes with a single thermal pad, which we've commented on, but as per the manual, should only be used on the rear port, not the topside heatsink. As I read it, the ports are numbered and the pad should only be used when installing an SSD on the rear port, so our comments here are justified.
2) We have to take everyone's overclocking abilities into account here. Some will only be happy playing with the voltage and multiplier at most, but even so, if a board is able to hit the maximum frequency of our CPU using our usual voltage with no other changes, then I think it deserves more credit/praise than a board that is potentially suffering from Vdroop depending on your setup, and will require extra tweaking. The fewer options you have to play with the better surely? I agree LLC is not a bad thing, but it's still an extra step you have to do that you don't on other boards. It's simple to do, but took time for us to investigate. I don't doubt many readers like that kind thing, but it still deserves to be mentioned and I would always look more favourably on a board that was able to reach 4.3GHz more easily than one that needed extra tweaking, especially there so many boards manage this, including previous mini-ITX generations compared to their ATX siblings. It's sometimes indicative of better power circuitry or additional headroom at the very least.
I also think it's absolutely fair to compare it not just to the Impact but ATX boards too. What you want from a mini-ITX board, especially one that costs more than £200, is one that competes with its bigger siblings in all areas except on I/O, ports etc. That's been our stance on enthusiast-focussed mini-ITX boards since the likes of DFI and Zotac set the scene over a decade ago, and rightly so, else we're not pushing manufacturers to deliver quality products and we'd slip back into dark old days of mini-ITX being, well, crap for anyone wanting one that's every bit as good as an ATX board, which most are these days. Also, we haven't said that these higher temperatures are an 'issue' and we don't expect anyone with have issues. That's why we gave the board an award. However, I would have concerns dropping in a 3950X and overclocking it.
3) We did post numbers for both VRMs and M.2/heatsink
But that's just it. All of the Impact's cooling is in the I/O shroud, it's used more/better fans, relocated the chipset - I feel Gigabyte could have done something similar. ASRock's competitor has a larger chipset fan that's attached to the VRM heatsinks via a heatpipe and from other reviews I've read seems to have cooler components as a result. Even with the Impact, Asus has only extended the board to fit the audio circuitry in flat and a bunch of extreme overclocking/watercooling features. Everything else is included within 170x170mm so I just feel that Gigabyte could have done more here and the better cooling absolutely made a difference, irrespective of the board being a bit longer, which is filled with components that the Gigabyte board mostly lacks anyway.
Only four sub ATX X570 boards exist.
The Gigabyte ITX
The ASRock ITX
The Asus DTX
The ASRock mATX
Bit-tech doesn't review ASRock products, so the Asus one was the only "competition" to choose from.
Actually that is incorrect there is an itx asus board as well as a dtx board.
I was more concerned that there was no mention of the price difference and size difference of the board they were directly comparing it to. No issue with them using what they have tested but lets just be clear on why the Gigabyte board is at a disadvantage compared to the asus, its far smaller and as such has more items crammed in.
Combatus I appreciate your input and apologies if my comments were coming across overly negative, as that was what I was trying to clear up from what I felt after reading you review of the Gigabyte
1) I missed that on my board when setting it up, didnt see that in the manual. Ive a thermal pad on during testing of the ssd in my case and temps never went about 45c during stress test, with average chipset temps of 65c under heavy load looped blender bmw. That could be the difference and I accept that you tested as was instructed so cant fault this on your part. Also cases etc have differences, I havent changed the chipset fan curve btw.
Oh i see them now, where does updated after the article originally published? Those values seem very high compared to what I have seen but glad they are in teh article
2) Like I said on my previous reply, I dont think you stating it is an issue more that the wording seemed negative as this was the focus rather than a stable overclock like the other baords. But I see where your coming from its more your method of reviewing.
3) No issue with comparing with atx/matx/dtx etc. however to say it should perform similarly from a vrm temps and chipset temps is a little mad considering the space constraints. Like for like vrms between atx and itx should perform similiar but when youve different phases etc. temps will differ and reduced heatsink sizes. Plenty of variables that can be attributed to size. Like how asrock splits power phases o two locations to reduce hot spots (downside to this is the reduced clearance and need for 115 socket on am4!!).
The dtx baord actually has the chipset moved if Im not mistaken as there is more space for a better cooling solution. I still think the itx baord from asus will run cooler than the gigabyte, I merely point out its not a simple apple to apples comparison.
Separate names with a comma.