Just upgraded to 7900X3D, can't tell the difference (not in games).

[kapone32]

No offense meant, but "it doesn't happen to meee" argument gets old really really fast

Yep you are the only person that know about the "true" AMD experience. That is why you argue with people that have AMD about how bad it is and reference experiences that are not based in anything other than conjecture. Like it is impossible for AMD drivers to be stable.

 

[Dr. Dro]

Yep you are the only person that know about the "true" AMD experience. That is why you argue with people that have AMD about how bad it is and reference experiences that are not based in anything other than conjecture. Like it is impossible for AMD drivers to be stable.

We've discussed this ad nauseam. There is nothing to add, this simply isn't going to be a productive debate, because you won't budge from your position regardless of what I say or present.

I've tried debating you in good faith before, it didn't work then and it won't work now. That, on top of being a thread hijack. I have far, far more experience than you ever gave me credit for, but the only end result here will be a "good for you." anyway.

 

[Vayra86]

CPU power has long passed the requirements of simple tasks on a PC. This isn't speculation, this is fact.

Heck, I could probably browse the web on a Sandy Bridge i5, and wouldn't notice much difference between that and my 7800X3D.

Im still running a 3570K to do basic stuff. Works fine...

 

[AusWolf]

With my laptop and the 6-7 generations' worth of AMD GPUs I had until I threw in the towel. No offense meant, but "it doesn't happen to meee" argument gets old really really fast

The same can be said from the other side, too, I'm afraid - the "my AMD card gives X error all the time" arguments get old, too.

I've had a shit ton of RDNA 2 and 3 GPUs, and never had any major problem with any of them. Sure, my experience with the 5700 XT was a bit of hit-and-miss (that's the last AMD GPU that gave me driver timeout errors), but ever since then, everything's been fine. I was also on AMD during the GCN era (HD 7000 series), and those cards were fine, too. I completely missed Polaris and Vega, so I can't comment on those.

Edit: The only major criticism I have towards AMD at this point is the abysmal video playback power consumption of RDNA 3 GPUs. I hope they'll fix it on RDNA 4.

 

[Zach_01]

CPU power has long passed the requirements of simple tasks on a PC. This isn't speculation, this is fact.

Heck, I could probably browse the web on a Sandy Bridge i5, and wouldn't notice much difference between that and my 7800X3D.

Its just that...
General CPU processing power is so high even a few gens back that we cant realize how high it is.
And the benchmarking today is focusing on very very specific scenarios to show CPU performance differences.
What most of us doing 80% of PC time can be done with 10y old hardware relatively easy.

Its like all of us driving a sub-5sec (0-60) cars and trying to find 0.1-0.2 sec differences.

 

[leonavis]

If an upgrade is to be useful, one need to identify potential bottlenecks for one's usecase.

For example, if you have a 4090 or 7900XTX and an Ryzen 5 2600 and want to play Baldur's Gate 3 or other CPU-intensive games, then surely your High-End-GPU will be bottlenecked by your rather old CPU, and quite severely.

However, playing very GPU-intensive games on high resolution, your CPU might - in most cases - even in this case not be the one harming performance. Of course, 1% lows should have gotten better even in your use-case, but general performance is very dependant on usecase and specific hardware. In most games, to the naked eye, there will be no difference visible between gaming on a 3700X and gaming on a 7800X3D, because it's not the CPU limiting the performance. And in your case you have upgraded from a top-of-the-line CPU to a, lets call it, topper-of-the-line CPU, so the differences in performance will be marginal at best, in most cases however non-existent.

Edit:

Is this to be expected with such tasks?

Oh, it's not about gaming.

yes, that is very much expected with such tasks.

Edit: The only major criticism I have towards AMD at this point is the abysmal video playback power consumption of RDNA 3 GPUs. I hope they'll fix it on RDNA 4.

Yeah, that's such a strange behavior. It cannot be that hard to get that under control. Quite annoying.

 

[AVATARAT]

Hey everyone,

I recently upgraded my rig from a 5900X, 3600 C16 RAM, and PCI-E 3.0 NVME to a 7900X3D, 6000 CL32 RAM, and Gen4 NVME.

So far, I’ve only had the chance to install some apps, browse the web, and do some light tasks, but I haven’t noticed any significant performance difference in these scenarios. Is this to be expected with such tasks?

For those of you who have upgraded from the 5000 series to the 7000 series, have you experienced similar results?

There is a difference, it all depends on how you use your computer and if you can fill the "magic". Some people can't understand because they don't pay attention to the details, because they don't care.
If you had faster RAM on AM4 and not as fast on AM5, it would mask some tasks you do.
So don't worry there is a difference

 

[AusWolf]

Its just that...
General CPU processing power is so high even a few gens back that we cant realize how high it is.
And the benchmarking today is focusing on very very specific scenarios to show CPU performance differences.
What most of us doing 80% of PC time can be done with 10y old hardware relatively easy.

Its like all of us driving a sub-5sec (0-60) cars and trying to find 0.1-0.2 sec differences.

Or like driving a sub-5sec car at a constant 30 mph (50 km/h).

 

[AVATARAT]

CPU power has long passed the requirements of simple tasks on a PC. This isn't speculation, this is fact.

Heck, I could probably browse the web on a Sandy Bridge i5, and wouldn't notice much difference between that and my 7800X3D.

Oh Sandy is so slow even when he's tuned in, please don't say it

Or like driving a sub-5sec car at a constant 30 mph (50 km/h).

Which is heavier, a kilogram of cotton or a kilogram of iron?
But nevertheless they are very different

 

[SL2]

Hahaha, I’m here, guys! Just didn’t have a chance to reply yet.

Oh, my GPU is a watercooled 3080 Ti.

First off, I’m not trolling anyone! I’m not expecting a massive performance gain, just that tiny bump in daily tasks – like installing new apps, extracting files, updating drivers, and copying files between NVMEs (I mean, 2.5GB/s from one KC3000 to another?).

When I upgraded from a Core 2 Duo E8500 to an E8600, yeah, I didn’t notice much. But moving from a Ryzen 1700 to a 2700X? Definitely felt it. And again from the 2700X to a 3700X, and then from the 3700X to the 5900X – each time, there was that little extra boost. Even if Gen3 vs Gen4 NVME isn’t night-and-day in real life, a faster boot time by even a second is something you notice!

Now I’ve gone from Gen3 to Gen4 NVME, DDR4 to DDR5, and a CPU that’s clocking an extra 600-700MHz more than my last one – so where’s that tiny boost?

Perhaps CPUs have just gotten so fast that these incremental gains aren’t noticeable in daily tasks anymore?

I’m not complaining, just trying to figure out why I didn’t get that extra little bump I was expecting.

That explains a lot!

I think a lot of people here jumped to the conclusion that you thought V-cache would make everything the fasterz, hence all the , but you're talking about generational uplift. Big difference lol.

 

[AusWolf]

Oh Sandy is so slow even when he's tuned in, please don't say it


Which is heavier, a kilogram of cotton or a kilogram of iron?
But nevertheless they are very different

I kindly disagree. I'm still convinced that there is (nearly) zero difference between a current gen, and a 10 year-old PC in web browsing.

 

[AVATARAT]

I kindly disagree. I'm still convinced that there is (nearly) zero difference between a current gen, and a 10 year-old PC in web browsing.

That's acceptable, after all a few seconds isn't much, but it's a few times slower than modern systems.

And I said it above, some people just don't care about the details. 0.5 or 3 seconds is a very big difference, but it ends up being short either way. On an untuned and unmaintenanced Sandy system it would open a page in 15-30 sec, for modern processors it would be slow but still a few times faster.

 

[mb194dc]

I kindly disagree. I'm still convinced that there is (nearly) zero difference between a current gen, and a 10 year-old PC in web browsing.

I have a 4790k based machine and a 5800x3d machine and can confirm. In general usage there's no difference between them.

The only usage there is a difference for me is in gaming and need better graphics card for 5800x3d to really show it I think outside of scenarios where I'm playing locally with a lot of bots.

 

[AusWolf]

That's acceptable, after all a few seconds isn't much, but it's a few times slower than modern systems.

And I said it above, some people just don't care about the details. 0.5 or 3 seconds is a very big difference, but it ends up being short either way. On an untuned and unmaintenanced Sandy system it would open a page in 15-30 sec, for modern processors it would be slow but still a few times faster.

Whether something is statistically different, or observably different is an entirely different story.

Our brains aren't mathematical devices. How many times your loading time is bigger or smaller doesn't matter as long as it's within acceptable boundaries.

 

[eidairaman1]

Browsing the web and light tasks and you want there to be a night and day difference? cool.... what GPU do you have, did you benchmark/ record gaming performance before upgrading, I mean, how fast does a web browser open these days on a potato you bought a gaming orientated performing CPU and say you can't notice any difference from your 12c pretty good previous CPU, you do know that once you go past pcie3 you won't notice a difference with pcie4/5 nvme in loading Windows or basic tasks unless benchmarking? we are at a really good point with computer HW these days as you can go back 4 years and still have a very fast and competant system running an AM4 CPU and DDR4 RAM, if you are seeking gaming performance going from AM4-AM5 and X3D AM5 then you better have a high end GPU to be CPU limited in most games, that said you should notice an uptick in .1% and .01% lows as well as minimums

Whats screwy is latencies continue to climb

 

[Frick]

With my laptop and the 6-7 generations' worth of AMD GPUs I had until I threw in the towel. No offense meant, but "it doesn't happen to meee" argument gets old really really fast

I will say that the only time the 6950xt has caused me problems is when I've dabbled in undervolting. I will also say that a driver for an AMD APU literally bricked a laptop (didn't even safe boot). I will also say that the specific "it doesn't happen to meee" argument works against you as well, because a lot of people have been running AMD GPU's and have them be stable. I will also say people have different ideas of "stable". The fifth thing I will say is that yeah sure AMD drivers are worse than Nvidia, with the addendum that, for me at least, it's not a big deal.

 

[A Computer Guy]

Hey everyone,

I recently upgraded my rig from a 5900X, 3600 C16 RAM, and PCI-E 3.0 NVME to a 7900X3D, 6000 CL32 RAM, and Gen4 NVME.

So far, I’ve only had the chance to install some apps, browse the web, and do some light tasks, but I haven’t noticed any significant performance difference in these scenarios. Is this to be expected with such tasks?

For those of you who have upgraded from the 5000 series to the 7000 series, have you experienced similar results?

Going from 5950x to 7950x I noticed things seem much snappier but I haven't tried playing any games on 7950x as it's used in my work machine.

 

[AsRock]

Hey everyone,

I recently upgraded my rig from a 5900X, 3600 C16 RAM, and PCI-E 3.0 NVME to a 7900X3D, 6000 CL32 RAM, and Gen4 NVME.

So far, I’ve only had the chance to install some apps, browse the web, and do some light tasks, but I haven’t noticed any significant performance difference in these scenarios. Is this to be expected with such tasks?

For those of you who have upgraded from the 5000 series to the 7000 series, have you experienced similar results?

What is your current v card ? as that might be the limiting factor.

 

[Fokker]

The only non-gaming tasks I could see those upgrades making a significant difference in is video editing, 3D render, or downloading/installing a large footprint program. Basic productivity tasks barely move the needle for high end hardware.

 

[SL2]

Spoiler

Yes, I'm repeating myself

The reading comprehension in this thread lol.

 

[GenericNinja]

The CPU has nothing to do with that drive transfer performance you are seeing. Per Techpowerup's review of the KC3000 in a large sequential transfer that drive starts at 5 GB/s and drops to 1.5 GB/s once the cache is filled, giving a final score of 2.0 GB/s. The 7 GB/s advertised for the KC3000 is the PCIE 4.0 x 4 interface speed, not what the SSD controller and NAND Flash are actually capable of achieving. Real world will always be far below that number, especially when files are small or a task requires random I/O.