M77-Mental (已封禁) 11 月 6 日 下午 8:15
Moved from 14600K to 7800X3D. My findings.
The 14600k had the same RAM I am using in the AM5 build at 7000mhz CL 32 and with a watercooler due to power consumption since watts = heat.
It was ran in a Z790 Livemixer motherboard.
I never had a single issue once using that system for a year.


7800X3D, it runs well, DDR5 at 6000mhz CL30, an air cooler, the Thermalright Royal Pretor 130.
It is in an MSI X870 Tomahawk Wifi and I have owned it for 2 days.
It is using all the same parts from the previous build except the platform of course.

So, what did I notice?

Demanding title, Borderlands 4, smoother on the 14600K, but less FPS by about 5%, the 7800X3D incurs more stutter but not enough to really complain much, it's just a bit more than I am used to.

BF6 runs the same as before, I was GPU limited at 1440P on my 7900 XTX, so the X3D shows no difference.

Moving on, I have notice a HUGE performance uplift with the 7800X3D whilst using Firefox which has privacy add-ons and is used in a strict custom config, it is vastly snappier and loads pages faster.

Both the 7800X3D and 14600K ship with integrated graphics, however the bonus for me is using an all Radeon driver instead of Intel and Radeon together, this stops issues such as Silent Hill 2 not loading on startup with the Intel iGPU enabled.


I did move all of this new platform into a new chassis, it is an FSP U500 workstation case, and the cooling is on a totally different level, 7900 XTX's are not known for having low hot spot temps, 80-85C is considered golden, this is running at 69C on maximum load with fans at 1500RPM on the card and the case fans at 1100RPM and near silent.

This is more a showcase of the case chosen than anything else, but it destroys my old Antec NX416L with all Noctua fans.

Images of system.

https://ibb.co/LhXTrWww

https://ibb.co/8gpK16jt

https://ibb.co/VYnLW1XJ

https://ibb.co/QvtfdZyT

And a video showcasing the 7900 XTX in that case with the 7800X3D.

https://www.youtube.com/watch?v=JfrTDlLg20g


This is my old Intel system.

https://www.elevenforum.com/attachments/488770-a5a4c07a75244817e4b3f7ca3f629976-1600-webp.137050/

https://www.elevenforum.com/attachments/488770-9e62e33439e64992c2c2669d050ffa57-1600-webp.137051/


https://www.elevenforum.com/attachments/488770-004ee799c93591bceef0906617b087b4-1600-webp.137053/



EDIT.

So I have benchmarked the 7800X3d in Cinebench R23, we score 18,133 CPU is stock.
14600K stock was doing 26,882

But it never felt like it took longer on the 7800X3D, CB runs are not slow anymore for modern CPU's so sure yes a 14600K is definitely A LOT faster in raw bench numbers, in reality though I didn't even notice the time slippage.

I don;t use my PC for work much so it's not much of a concern but in all fairness the 7800X3D and 14600K are essentially neck and neck in all round usage so far.

7800X3D will score higher in games, 14600K will score higher in work loads.
Keep in mind the AM5 system has a memory speed limitation for latency and to get the lowest latency the CPU has to run the RAM at 6000, you can try higher but 6000 is the sweet spot and highest potential.

The Intel has vastly higher bandwidth and lower latency by almost 50% on total memory bandwidth and about 20% lower latency, but that dos not translate to games unless you were pitting it against non X3D Ryzen CPU's. It was using the RAM to it's full potential at 6800mhz CL32 with tight tertiaries.

Neither CPU's weakness is a real weakness, they only look that way to internet fanboys.


EDIT updated with CS2 casual 10 vs 10 gameplay.

https://www.youtube.com/watch?v=1Qzwly1Yi1w
最后由 M77-Mental 编辑于; 11 月 9 日 上午 10:50
< >
正在显示第 1 - 15 条,共 53 条留言
smallcat 11 月 6 日 下午 8:22 
RTX 5090 would benefit from such a change . Your card is fast but i5 14600K can handle it , especially at 1440p and 4K
https://www.tomshardware.com/reviews/cpu-hierarchy,4312.html
最后由 smallcat 编辑于; 11 月 6 日 下午 8:36
In no case would a 14600k offer more performance or fluidity especially when the 30% worse 1% and .1% lows.
smallcat 11 月 6 日 下午 9:01 
引用自 The_Abortionator
In no case would a 14600k offer more performance or fluidity especially when the 30% worse 1% and .1% lows.
not true
https://www.techpowerup.com/review/intel-core-ultra-9-285k/22.html
M77-Mental (已封禁) 11 月 6 日 下午 9:03 
引用自 The_Abortionator
In no case would a 14600k offer more performance or fluidity especially when the 30% worse 1% and .1% lows.
You seem to be misinformed, must be a lack of experience and ownership.

Only in BL4 did the 7800X3D perform slightly worse in fluidity, it's not a knock, just reporting facts for what I have observed.
最后由 M77-Mental 编辑于; 11 月 6 日 下午 9:04
引用自 smallcat
引用自 The_Abortionator
In no case would a 14600k offer more performance or fluidity especially when the 30% worse 1% and .1% lows.
not true
https://www.techpowerup.com/review/intel-core-ultra-9-285k/22.html

Those aggregate sites aren't accurate. You really should look at real benchmarks instead.
smallcat 11 月 6 日 下午 9:50 
The_Abortionator , the best source provided . i have seen tech guys who dont know what polling rate of mouse is . Dont saying for you . Moreover you havent tested so much .
Monk 11 月 6 日 下午 9:58 
Benchmarks should not be used to judge fluidity and smoothness, actual game play should and despite having lower 1 and 0.1% lows, they do not happen as often nor is the drop as severe usually.

There is a reason I'll suggest 9000 series but wouldn't previous gens, they simply offer an inferior gaming experience overall, 9000 still has more frequent drops in any game play you can go look at, but it's no where near as bad.
引用自 smallcat
The_Abortionator , the best source provided . i have seen tech guys who dont know what polling rate of mouse is . Dont saying for you . Moreover you havent tested so much .

That is far from the best source provided period. Also what does mouse polling rate have to do with any of this?

That source sucks.
引用自 Monk
Benchmarks should not be used to judge fluidity and smoothness, actual game play should and despite having lower 1 and 0.1% lows, they do not happen as often nor is the drop as severe usually.

There is a reason I'll suggest 9000 series but wouldn't previous gens, they simply offer an inferior gaming experience overall, 9000 still has more frequent drops in any game play you can go look at, but it's no where near as bad.


You are sounding exactly like audiophiles. Ignoring facts and scientific data and substituting it with emotions is meaningless.

When what you claim cant be measured and proven its because you're wrong.
Dom 11 月 7 日 上午 12:15 
Most of these game tests don't really say which CPU is actually better but which CPU's iGPU is, if that's what you're using. Which is a minor detail when you consider everything CPUs are used for and also the fact that almost every gaming build has dedicated graphics cards.

People don't always realize that worse CPUs may have better built-in graphics.
最后由 Dom 编辑于; 11 月 7 日 上午 12:18
ChickenBalls 11 月 7 日 上午 2:08 
引用自 HissingPython
引用自 The_Abortionator
In no case would a 14600k offer more performance or fluidity especially when the 30% worse 1% and .1% lows.
You seem to be misinformed, must be a lack of experience and ownership.

Only in BL4 did the 7800X3D perform slightly worse in fluidity, it's not a knock, just reporting facts for what I have observed.

yeah but BL4 is not known to be an optimized game
kind of like Starfield, they seem to run better on Intel cpus for some strange reason while all the other newer games run better on AMDs X3D cpus
_I_ 11 月 7 日 上午 2:33 
引用自 Dom
Most of these game tests don't really say which CPU is actually better but which CPU's iGPU is, if that's what you're using. Which is a minor detail when you consider everything CPUs are used for and also the fact that almost every gaming build has dedicated graphics cards.

People don't always realize that worse CPUs may have better built-in graphics.
op has 7900xtx

its impossible to see cpu bottlenecks when the gpu is the limiting factor
Dom 11 月 7 日 上午 3:02 
引用自 _I_
引用自 Dom
Most of these game tests don't really say which CPU is actually better but which CPU's iGPU is, if that's what you're using. Which is a minor detail when you consider everything CPUs are used for and also the fact that almost every gaming build has dedicated graphics cards.

People don't always realize that worse CPUs may have better built-in graphics.
op has 7900xtx

its impossible to see cpu bottlenecks when the gpu is the limiting factor
Pretty much, built-in graphics are by themselves the "bottleneck" here.

You may have a very powerful CPU that renders content amazingly, great single-core & multi-core performance, but it's just bad at graphics because that's not where the chip's design focus is. Integrated graphics are generally made for basic use, streaming videos (2D). They're not made for 3D rendering and games.

And because integrated graphics are not so much focused on, they can be better in a chip that has worse performance for those tasks CPUs are typically used for.

I mean why would chip manufacturers even focus so much on integrated graphics when most computer users are going to have dedicated graphics cards anyway?
最后由 Dom 编辑于; 11 月 7 日 上午 3:04
Slav Mcgopnik 11 月 7 日 上午 4:54 
引用自 The_Abortionator
引用自 Monk
Benchmarks should not be used to judge fluidity and smoothness, actual game play should and despite having lower 1 and 0.1% lows, they do not happen as often nor is the drop as severe usually.

There is a reason I'll suggest 9000 series but wouldn't previous gens, they simply offer an inferior gaming experience overall, 9000 still has more frequent drops in any game play you can go look at, but it's no where near as bad.


You are sounding exactly like audiophiles. Ignoring facts and scientific data and substituting it with emotions is meaningless.

When what you claim cant be measured and proven its because you're wrong.
Yeah it’s one of Monk’s quirks. Definitely knowledgeable about PCs but is hung up on this supposed undetectable instability in AMD CPUs (probably because they are insecure over Intel not being better and have to use the Framechasers cope if “Intel is better for enthusiasts”).
Monk 11 月 7 日 上午 5:15 
引用自 Slav Mcgopnik
引用自 The_Abortionator


You are sounding exactly like audiophiles. Ignoring facts and scientific data and substituting it with emotions is meaningless.

When what you claim cant be measured and proven its because you're wrong.
Yeah it’s one of Monk’s quirks. Definitely knowledgeable about PCs but is hung up on this supposed undetectable instability in AMD CPUs (probably because they are insecure over Intel not being better and have to use the Framechasers cope if “Intel is better for enthusiasts”).

It's viewable in most open world game plays online you can go watch, all systems will have drops, it's just something that has existed since ryzen was released due to the way the architecture works.

As to needing to think Intel is better or amd, I don't care who makes the chip I use or recommend, I always and for longer than some have been alive I suspect, suggest the option that gives the best experience for the individual use case, for a long time that was clearly Intel, with the 9000 series amd have taken the lead and their one weakness is drastically reduced.

See, unlike you fanboys I can accept and recognise flaws in the various options regardless of who makes it.

I mean I am such an Intel fanboy that I tell 99.9% of people to buy amd over Intel, but recognising a flaw in your beloved brand seems to be hiretical to some of you.

If you have never noticed it and don't care, that's fine, but it doesn't mean that it doesn't happen any more than people saying their Intel chips didn't degrade doesn't mean they don't because it was a flaw in the design, only with ryzen over 5 generations and a decade they have polished it enough that it's barely there and the upsides outweigh the one (now small) negative.

The only ones who seem to cope are those who put their head in the sand and pretend amd never had lattency and stutter issues, these companies don't care about you, it's OK to admit that things have flaws, heck, I only have my current setup due to a convuluted series of circumstances, if I was building from scratch I'd of gone with a 9800x3d myself, but I have a 14900ks and slightly smoother game play for it, all be it at a lower fps.
< >
正在显示第 1 - 15 条,共 53 条留言
每页显示数: 1530 50