WTF??? You were serious, there's a 2300X CPU? Where would you find one of these, never seen it on Newegg.
WTF??? You were serious, there's a 2300X CPU? Where would you find one of these, never seen it on Newegg.AMD 2300x vs intel 2500k.
https://www.anandtech.com/bench/product/2266?vs=2355
The 2300x has less than a 10% clock advantage but smokes the 2500k in most tests. Both are 4 cores and 4 threads.
WTF??? You were serious, there's a 2300X CPU? Where would you find one of these, never seen it on Newegg.
Here you go:I'm not interested in starting an argument, but when I looked online I found several sources showing the 4 core Ryzen chips right in line with the 2500K. If an R5 1400 with more cache and SMT cant beat a 2500k at similar clocks, well that says to me they perform very similar clock for clock. Yes a 2700 is double that, but for a bunch of single threaded apps none of that will matter (it should simply allow more of those apps to run at once)...and if the 2700 is running at a 20% lower speed than a 2500K then I can reasonably assume it's going to be slower for each app (and real world usage backs that up).
I will get around to further experiments on the new system, but it wont be for a while. Considering how much time, effort, and money I already put into it, my disappointment can easily turn to disgust and hatred. I don't want/need that kind of crap.
Sell it to someone that can make it work then if you can't. They are pretty user friendly, I'm sure that you'll find a buyer.I'm not interested in starting an argument, but when I looked online I found several sources showing the 4 core Ryzen chips right in line with the 2500K. If an R5 1400 with more cache and SMT cant beat a 2500k at similar clocks, well that says to me they perform very similar clock for clock. Yes a 2700 is double that, but for a bunch of single threaded apps none of that will matter (it should simply allow more of those apps to run at once)...and if the 2700 is running at a 20% lower speed than a 2500K then I can reasonably assume it's going to be slower for each app (and real world usage backs that up).
I will get around to further experiments on the new system, but it wont be for a while. Considering how much time, effort, and money I already put into it, my disappointment can easily turn to disgust and hatred. I don't want/need that kind of crap.
Man, just run cinebench R15 or R20 and share screenshots of cpuz and gpuz with us. There is no way 2500K is anywhere near Ryzen 2700.I'm not interested in starting an argument, but when I looked online I found several sources showing the 4 core Ryzen chips right in line with the 2500K. If an R5 1400 with more cache and SMT cant beat a 2500k at similar clocks, well that says to me they perform very similar clock for clock. Yes a 2700 is double that, but for a bunch of single threaded apps none of that will matter (it should simply allow more of those apps to run at once)...and if the 2700 is running at a 20% lower speed than a 2500K then I can reasonably assume it's going to be slower for each app (and real world usage backs that up).
I will get around to further experiments on the new system, but it wont be for a while. Considering how much time, effort, and money I already put into it, my disappointment can easily turn to disgust and hatred. I don't want/need that kind of crap.
If thats the case, then some benchmarking can confirm it.Is it possible, that you fell prey to a "fake CPU"? If you bought it on ebay, and especially if you bought it from a China seller, rather than an "authorized" reseller (like Newegg, Platinum Micro, or BestBuy), then your CPU might be fake.
You might ask, how can you fake a CPU? Well, there's faked heatspreaders, and then, there can be faked micro-code or CPUID strings, stuff like that. Maybe it's actually a Bristol Ridge CPU.
Unlikely, of course, but I thought that I would try to help cover all of the bases here on this subject.
Understanding the different CPU frequency states (PStates), their voltages and especially the actual effective voltage is harder than ever before with Zeppelin. Unlike with the older designs (15h family) the boosted PStates (Turbo & XFR) are completely invisible.
Due to that fact, they are officially called as "Shadow PStates". This means that unlike with the previous designs these PStates are not defined in the standard MSR registers and cannot be modified (or be seen) by the user. The only way the user can even verify their presence is to see them actually firing (i.e. from the actual effective frequency & voltage).
This is matching up perfectly with all the things he said...Here you go:
https://www.hardware.fr/articles/965-2/performances-applicatives.html
baseline is 2500K, very handy to compare to other chips. R3 1200 (lower clocks than 2500K, no SMT) is 1% ahead of 2500K in application performance (aggregate of all tests). R5 1400 is 35% ahead.
As can be seen by above what you have in your machine is not matching up with results from reputable sources. Just wanted to bring this up to clear the fud in the topic.
Yup, you can scale your results to ~4Ghz for the ryzen and ~5Ghz for the 2500k to know what a fully O/C system for both sides will look like and the 2500k would end up about 20% faster per core.
This is matching up perfectly with all the things he said...
"when I looked online I found several sources showing the 4 core Ryzen chips right in line with the 2500K. If an R5 1400 with more cache and SMT cant beat a 2500k at similar clocks, well that says to me they perform very similar clock for clock. "
Similar clock sameish performance ... check
"Yes a 2700 is double that, but for a bunch of single threaded apps none of that will matter (it should simply allow more of those apps to run at once)... "
Well that's just common sense.
"and if the 2700 is running at a 20% lower speed than a 2500K then I can reasonably assume it's going to be slower for each app (and real world usage backs that up). "
Yup, you can scale your results to ~4Ghz for the ryzen and ~5Ghz for the 2500k to know what a fully O/C system for both sides will look like and the 2500k would end up about 20% faster per core.
There are many new instruction sets which are available on ryzen but not on sandy bridge 2500k. For ex AVX2. Even if 2500k @ 5ghz, it wont compete with any 2nd gen ryzen in IPC. Memory performance (dual ch ddr3 1600mhz on 2500k vs dual ch ddr4 3200mhz on ryzen) and L3 cache bandwidth performance can also create big difference in many apps. Ryzen 5 1400 @ 4.1ghz scores ~920CB = i7 6700k @ 4.4ghz. I5 2500k @ 4.4ghz scores ~ 560CB.This is matching up perfectly with all the things he said...
"when I looked online I found several sources showing the 4 core Ryzen chips right in line with the 2500K. If an R5 1400 with more cache and SMT cant beat a 2500k at similar clocks, well that says to me they perform very similar clock for clock. "
Similar clock sameish performance ... check
"Yes a 2700 is double that, but for a bunch of single threaded apps none of that will matter (it should simply allow more of those apps to run at once)... "
Well that's just common sense.
"and if the 2700 is running at a 20% lower speed than a 2500K then I can reasonably assume it's going to be slower for each app (and real world usage backs that up). "
Yup, you can scale your results to ~4Ghz for the ryzen and ~5Ghz for the 2500k to know what a fully O/C system for both sides will look like and the 2500k would end up about 20% faster per core.
There are several versions of the 212 these days but I believe all of them now are am4 compatible or there is a adapter you can buy seperately.
Ryzen+ is Broadwell level of IPC. There is no way that SB can be 20% faster per core if it's running at ~25% higher clockspeed. Unless you're suggesting that Broadwell level of IPC is just 5% over SB.This is matching up perfectly with all the things he said...
"when I looked online I found several sources showing the 4 core Ryzen chips right in line with the 2500K. If an R5 1400 with more cache and SMT cant beat a 2500k at similar clocks, well that says to me they perform very similar clock for clock. "
Similar clock sameish performance ... check
"Yes a 2700 is double that, but for a bunch of single threaded apps none of that will matter (it should simply allow more of those apps to run at once)... "
Well that's just common sense.
"and if the 2700 is running at a 20% lower speed than a 2500K then I can reasonably assume it's going to be slower for each app (and real world usage backs that up). "
Yup, you can scale your results to ~4Ghz for the ryzen and ~5Ghz for the 2500k to know what a fully O/C system for both sides will look like and the 2500k would end up about 20% faster per core.
But if he's not running programs that need more than 4 cores, why did he buy a 2700 in the first place? Why go for a chip you know isn't right for your use case?
*snip*
Yeah, I looked it up. There's the 212 led turbo that comes with 2 fans, but the evo isn't compatible. You can buy a separate compatibility kit but they seem to be out of stock everywhere.
It seemed logical that a CPU equivalent to twice the current one would handle the large pool of software better (and more efficiently utilize RAM).
"problem".
However I'm happy to admit there are some things I can still try, if I can get it to perform better obviously that's a good thing.
Decided what the heck, I'll log in and harass some people
I mean you're not even hitting XFR frequencies so something is definitely wrong in your setup.