• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD THREADRIPPER VS INTEL SKYLAKE X

Routers aside I've never had great results with Asus - over the years (about 25 boards going back just over 16 years) their motherboards have far more often degraded over time than any other brand I've used i.e. DIMM slots stop working, LAN ports malfunction, chipset ICs come loose of their soldering, etc. also not had a great track record with their monitors having been through 2 of the ROG Swift that have died and know people IRL who've been through several before giving up and buying another brand.

funny I've got an asus router, ended up binning it due to constant Wi-Fi drops and it kept giving my strict NAT in games, something even my ****box virgin superhub manages fine.

but I'm tempted to go either gigabyte or msi this time around, never had an issues with my gaming x gpus, never tried a gigabyte product so not sure which to try next. evga would be an option but they're incredibly slow these days to get products to market.

And then you see second generation Ryzen boards like the cheap Strixx beating the CH6 and having the same performance with the MSI Titanium.

how do you mean 'beating' ?
 
funny I've got an asus router, ended up binning it due to constant Wi-Fi drops and it kept giving my strict NAT in games, something even my ****box virgin superhub manages fine.

but I'm tempted to go either gigabyte or msi this time around, never had an issues with my gaming x gpus, never tried a gigabyte product so not sure which to try next. evga would be an option but they're incredibly slow these days to get products to market.

I've not seen much in the way of hardware failure/degradation over time though with the routers unlike other hardware they make - and most of the software issues can be solved by flashing them with Merlin's firmwares.
 
I've not seen much in the way of hardware failure/degradation over time though with the routers unlike other hardware they make - and most of the software issues can be solved by flashing them with Merlin's firmwares.

ah so the routers are rebranded merlins? wonder if that would fix my one....
 
No different Merlin - some guy that has been creating custom firmwares for a variety of routers for awhile - his custom firmwares for the popular Asus models almost entirely fix for the most part the sloppy work in the retail software.

Asuswrt-Merlin is an alternative, customized version of that firmware. Developed by Eric Sauvageau, its primary goals are to enhance the existing firmware without bringing any radical changes, and to fix some of the known issues and limitations,
 
I'll keep that in mind. No Asus motherboards.
What are you guys on ASUS is pretty much the only brand doing solid MB engineering and testing. At the NDA event for X299 they was way ahead of the competition.

We used hundreds of WS boards almost without issues.
 
What are you guys on ASUS is pretty much the only brand doing solid MB engineering and testing. At the NDA event for X299 they was way ahead of the competition.

We used hundreds of WS boards almost without issues.

How long do you keep a motherboard for?
 
The last bit of his post applies to long term experience.

Sadly as a consumer and occasional system builder my experience long term hasn't been great (I didn't take a hammer to one of Asus's higher end boards for no reason heh), maybe I've been unlucky but brands like Gigabyte have worked out far better for me.
 
Used a few ASUS boards in the past and were fine, but that's some time ago. I had an ASUS router that was sh** though(ASDL). I though it was just poor quality line at a place I was renting and when I moved into my own place and bought a TP-LINK VDSL for fibre router I sold the ASUS one to a forum member who couldn't even get it working with their boardband supplier. Had to take it back and still sits in a box in the attick as a spare.
Oh and this weekend the TP-LINK router died too, well, suddenly stopped allowing internet traffic and simply won't work. Wifi works but no internet Rebooted it, set back to factory settings but nope -not even with help from Zen could we get it working. Stil loads on the warranty but t's one of those things you cannot do without so had to buy a new one. Have little time for RMA so that sh** unit will be binned.
Returned to a Netgear router now......
Back on topic I always though ASUS boards were the best. Must admit the Gigabyte Ive used for 5 years now has been awesome and the MSI board before that also pretty good.
 
What are you guys on ASUS is pretty much the only brand doing solid MB engineering and testing. At the NDA event for X299 they was way ahead of the competition.

We used hundreds of WS boards almost without issues.
TBH, if ASUS is way ahead of the competition and they currently can't get stock voltages right. Its kind of worrying.
 
Sold one of my cars so I've got £2500 for an upgrade - Cmon thread ripper and vega, show me what you've got before I spend it all!
 
This is worth watching, it explains Slylake-X performance oddities, and points out some PcPer review shenanigans. IE they are setting 7.zip to use only 8 threads which makes lower threaded CPU like the 7700K much better than they actually are compared with CPU's like the 1800X and they are testing compression rather than decompression which again makes Intel look better in comparison than they would using decompression.

 
IE they are setting 7.zip to use only 8 threads which makes lower threaded CPU like the 7700K much better than they actually

Would need testing to know if it is still the case but a few years back you actually got slower performance with 7z beyond a certain number of thread counts even to the point sometimes of taking longer to complete than 1 thread - under those conditions the rate of slow down on the lower core count CPUs was even more than the big ones. I believe this was partly due to the bottleneck being IO rather than CPU and increasing the access requirements over lots of threads just added queue bottlenecks on top of the already constrained IO - again though I have no idea if that is still true today as its been a good few years since I looked at 7z in any detail.
 
Would need testing to know if it is still the case but a few years back you actually got slower performance with 7z beyond a certain number of thread counts even to the point sometimes of taking longer to complete than 1 thread - under those conditions the rate of slow down on the lower core count CPUs was even more than the big ones. I believe this was partly due to the bottleneck being IO rather than CPU and increasing the access requirements over lots of threads just added queue bottlenecks on top of the already constrained IO - again though I have no idea if that is still true today as its been a good few years since I looked at 7z in any detail.

Adored actually ran 7Zip bench on 8 and 16 threads for Ryzen and got improved results. A 50% increase in performance for decompression as well.
 
Ah cool haven't watched the video yet - from skipping near the end seems PCPer limited to 8 threads so they could recycle old results instead of re-testing everything.
 
Ah cool haven't watched the video yet - from skipping near the end seems PCPer limited to 8 threads so they could recycle old results instead of re-testing everything.

Its why they went into the settings to limit the application to 8 threads to start with that gets me, the only reason you would do that is to stop anything with more than 8 threads getting ahead of 8 thread CPU's.

Like i said, PcPer review shenanigans......
 
Its why they went into the settings to limit the application to 8 threads to start with that gets me, the only reason you would do that is to stop anything with more than 8 threads getting ahead of 8 thread CPU's.

Like i said, PcPer review shenanigans......

In the video he claims it is because they originally changed to the benchmark when CPUs only went to 8 threads (atleast the ones relevant to their testing at the time) and suggests they didn't want to re-test all the CPUs again - as he points out they are recycling old results. Ultimately a bit meaningless without atleast testing the newer ones at their full potential.
 
In the video he claims it is because they originally changed to the benchmark when CPUs only went to 8 threads (atleast the ones relevant to their testing at the time) and suggests they didn't want to re-test all the CPUs again - as he points out they are recycling old results. Ultimately a bit meaningless without atleast testing the newer ones at their full potential.

Well, for one 8 thread CPU's have been around for a decade, thats a decade before he started using 7.zip to review CPU's.
Two, no one ever changes how many threads to use, it just uses what threads are available, you would only set it if you wanted to use the set number of threads and no more, he did this reviewing the 8 thread 7700K against the 16 thread 1800X, in other words he set it so 7.zip would not use more threads on the 1800X than the 7700K had, you would only do that if you didn't want the 1800X to result higher performance than the 7700K.

PcPer review shenanigans......
 
Back
Top Bottom