Also this just released:
Did you just assess system stability of a 5600X @ 4.6Ghz / 1.2V with just a couple of CB 15 runs?Sad to see PBO being as retarded and useless as it always was. Pumping 1.3V for negative gains in CB15 score @4.6Ghz sounds just what brain dead OC would do.
Did you just assess system stability of a 5600X @ 4.6Ghz / 1.2V with just a couple of CB 15 runs?
Sad to see PBO being as retarded and useless as it always was. Pumping 1.3V for negative gains in CB15 score @4.6Ghz sounds just what brain dead OC would do.
Those who care about efficiency will still need to run static all core overclocks, loosing ST peak clocks it seems.
Same here on my 3900X. Static overclocks are lame to (4.2Ghz all core to be stable under all conditions). Undervolting helped a bit until a BIOS update.Yup. On my 3700X, PBO gets me 4.4Ghz peak at ~1.4V (HWInfo has shown peaks around 1.45!), temps peak around 78C on water. I know part of that high temp is the chiplet design and not having a Ryzen optimized block but still. Manual overclock is 4.2Ghz at 1.14V with temps around 60C. PBO has been utter trash for my sample.
Yup. On my 3700X, PBO gets me 4.4Ghz peak at ~1.4V (HWInfo has shown peaks around 1.45!), temps peak around 78C on water. I know part of that high temp is the chiplet design and not having a Ryzen optimized block but still. Manual overclock is 4.2Ghz at 1.14V with temps around 60C. PBO has been utter trash for my sample.
Yes. On air. There is a video about it here:
Yup. On my 3700X, PBO gets me 4.4Ghz peak at ~1.4V (HWInfo has shown peaks around 1.45!), temps peak around 78C on water. I know part of that high temp is the chiplet design and not having a Ryzen optimized block but still. Manual overclock is 4.2Ghz at 1.14V with temps around 60C. PBO has been utter trash for my sample.
Do you have by any chance a list like this for Linux systems?Geekbench 5.2 Single Threaded CPU scores:
(So this is still not Geekbench version 5.3 using VEAS which should increase the Zen 3 scores significantly.
View attachment 32809
.
I'm tempted to start asking people about their stock voltages on Intel Skylake based CPUs. Then we go move on to Nvidia's Ampere for a complete tour.
What if I told you that Intel platforms also easily push 0.1V more than they should? What if we started to look at stock Vcore based on motherboard model / BIOS version?
Any bets how long it takes userbenchmark to add tweaks specific to Ryzen 5000?
Do you have by any chance a list like this for Linux systems?
no, not at all. rather, it seems to me more that tech writers, who have been in serious need of line editing for as long as i can remember, didn't know that a suitable form of the word "perform" already existed, and so invented and started using "performant." without even being aware that there is an actual french word.It's actually a word, albeit a French word. It would appear that it's in the process of being incorporated into the English language (which historically has borrowed a lot of words from other languages, French especially) and that the usage in English is essentially the same meaning that it has in French. Otherwise, it's a perfectly cromulent word.
If it being French makes you think that means it's somehow not an English word, then obviously this is your first rodeo.
no, not at all. rather, it seems to me more that tech writers, who have been in serious need of line editing for as long as i can remember, didn't know that a suitable form of the word "perform" already existed, and so invented and started using "performant." without even being aware that there is an actual french word.
again, see second comment:
Intel Xeon 3.6 2MB vs AMD Opteron 252 Database Test
www.anandtech.com
But...but you have forgetten every ++ after 14nm also counts for additional performance tweaks which shall be added to the scoreI think they're running out of tweaks they can make at this point. About the only way Intel comes out ahead is if they benchmark the number of letters in the manufacture's name in which case Intel has a comfortable 66% lead over AMD and if we look specifically at vowels than Intel has doubled up their performance lead in that category.
The number in the process node is also about 42% bigger for Intel as well so I guess they've got an edge over AMD in that regard as well. If they showed a graph for "financial considerations" I have no doubt that Intel would come out ahead there too!
no more assumption than you've displayed.That's a lot of assumption on your part about the minds of people who are not you. Can you definitively say that neither of the authors of that article have never spoken French or taken a course on the language? I doubt it.
Even if you could, it doesn't matter. They could have simply heard someone who speaks French use the word. The important part of language is that we understand what is meant to be conveyed. Is anyone confused or mislead by the use of "performant" in a sentence? That it's easy for anyone to grok from basic context causes it to enter the lexicon and become used by others to convey meaning.
And my point is people are comparing PBO voltages versus manually tuned voltages. One can easily get 100+mV undervolts even starting from stock settings, on all platforms.Think the point is that PBO pushes voltages up when there isn't necessarily a need to in order for another 100Mhz or so.
Benchmarks 5800x
System manufacturer System Product Name - Geekbench
Benchmark results for a System manufacturer System Product Name with an AMD Ryzen 7 5800X processor.browser.geekbench.comSystem manufacturer System Product Name - Geekbench
Benchmark results for a System manufacturer System Product Name with an AMD Ryzen 7 5800X processor.browser.geekbench.com
C20: 5616
Normal 3200 c16 ram
Yes. On air. There is a video about it here: