Soldato
- Joined
- 22 Nov 2018
- Posts
- 2,715
B350 and X370 can't. B450 and X470 are subject to the board partners but most will be supported.
-
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 Zen 2 (Refresh) 3900XT/3800XT/3600XT
B350 and X370 can't. B450 and X470 are subject to the board partners but most will be supported.
Not all AM4 boards will support Zen3 so the XT chip might hold some value. Ie the XT range will be the fastest supported and last DDR4 chips.
It looks like some 300 series boards might support Zen 3
i thought AMD said no 4 series could, because of the size of the bios?
Some?
Where did you see that? Have 1600 and b350 myself, would prob get a 3700x or 3800xt at some stage if I cant get the 4 series
I’d only read the original story. 4 series is go now
Soldato
What I posted makes perfect sense. Now this right here makes no sense...
Really? So the jump from a 2600X to 3800XT is minimal? The jump from my i5 2500 to 3800XT is minimal? lol OK
There's more to an upgrade than just a single generation, and even then you can see big gains between single generations that make such an upgrade worth it.
Not sure if this has been posted but I consider it potentially big.
Some motherboard vendors have been cheating on power consumption reported to the chip to manipulate XFR. In some cases it can be as high as 50% power usage not been reported, I dont know if this affects whats reported in power monitoring software, but my guess is it does. Testing shows in most cases the clock speed gain is a mere 100mhz, the lengths these board vendors go to for small levels of performance is getting out of hand. But whats nasty about this one, with the exception of MSI who make it tuneable in the bios, its a hidden tweak, that the end user is not aware of and has no means to fix.
https://www.techpowerup.com/268260/...ow-motherboard-makers-are-cheating-ryzen-cpus
Some motherboard vendors have been cheating on power consumption reported to the chip to manipulate XFR. In some cases it can be as high as 50% power usage not been reported, I dont know if this affects whats reported in power monitoring software, but my guess is it does. Testing shows in most cases the clock speed gain is a mere 100mhz, the lengths these board vendors go to for small levels of performance is getting out of hand. But whats nasty about this one, with the exception of MSI who make it tuneable in the bios, its a hidden tweak, that the end user is not aware of and has no means to fix.
https://www.techpowerup.com/268260/...ow-motherboard-makers-are-cheating-ryzen-cpus
I see I am going to have to spell things out a little more clearly if you are having trouble with this. There is no ponit in buying a XT CPU when the non-XT CPU's are so close in performance. It makes more sense to upgrade from an i5 2500 to a 3600 or 3700x, save a fair bit of cash and get similar performance. The XT CPU's are not worth the extra money over their predecessory for the performance increase they provide.
Is that clearer now? :?
Soldato
Oh, so I'm unclear on an opinion you have? I'm sorry, I thought you were talking about actual facts. Care to share your benchmarks and official pricing then? Because clearly you have such information to be so absolute in your statement.
You may well be proven correct when the XTs land, or you may not. We may see other factors come into play such as lower power draw or better thermals coming from mature silicon. 2GHz Infinity Fabric may net some surprising gains, worth any price differential that may exist.
You may well be proven correct when the XTs land, or you may not. We may see other factors come into play such as lower power draw or better thermals coming from mature silicon. 2GHz Infinity Fabric may net some surprising gains, worth any price differential that may exist.
Caporegime
The thread they link to on the hwinfo forums says things are being blown out of proportion.
No problem, they are facts based on the things we know about the CPU's presented by AMD themselves. But I have been on forums way too long to expect common sense from everyone so I am happy to agree to disagree. To summarize for me to refer back to later:
- Performance increase of XT's will be relatively small, we obviously already see this through people overclocking 3600's/3800's/3900's since months so it's a no-brainer as XT's will just be higher clocked (by 200-300mhz) versions of existing CPU's.
- Prices of 3600/3700/3800/3900 will drop to make way for the refresh (they are already dropping in the USA). The 3600 (non-x) can now be had for $165 which is insane value for even high-end gaming and will anyway overclock to around XT speeds with a passable mobo/RAM. 40% less for a CPU that performs largely the same in all meaningful real-world scenarios? Crazy.
Yeah, seems like a bit of a non-issue in real terms.
Last edited:
https://www.youtube.com/watch?v=10b8CS7wQcM
Some number and qualifications there, seems less impactful than MCE tweaks on Intel.
Some number and qualifications there, seems less impactful than MCE tweaks on Intel.
The more I read about this the more I absolutely do not think it is important. Seems like a sensationalist article that was written by Tomshardware with a hysterical response from some of the community.
I'd assume the boards that do it have VRM's that are beyond standard specification. As long as cooling is sufficient it shouldn't cause a problem unless it causes the boost feature to use dangerous levels of voltage? you're more likely to get damage to VRM's and I don't think motherboard vendors will do anything to make their boards unreliable.
Who are the two vendors doing it? Asus and Gigabyte?
Who are the two vendors doing it? Asus and Gigabyte?
Last edited:
Sounds like something to stir up a slow news day. The main scummy thing is motherboard manufacturers apparently boosting it during launch day bios but not later daily drivers.
Yeah but hardware manufacturers have been doing petty tricks like this in various forms for years. In the end the impact is negligible and it comes more down to principle.
