I am a hypocrite, but there is really not that much need to upgrade CPUs anymore. Even a ten year old chip seems completely adequate for day to day use. I played with a N100 recently and those things are incredibly capable.

(Ignore my AM5 workstation with 192GB RAM in the corner)

I rocked my Haswell i5 until last year when I built a brand new machine around the 9800x3d. Along the way I upgraded it from 8gb of ram to 32gb, got a gen 1 pcie3 NVME, and went through successive hand-me-down GPUs starting from a GeForce 770 to the RTX 2070 it has now.

In fact my wife is still rocking that machine - although her gaming needs are much less equipment intense than mine. After a small refurb I gave it (new case, new air cooler, new PSU) - I expect it to last another 5 years for her.

I rode out an i7-4790K until this year... replaced solely because of Windows 10 support ending. But it's a solid chip.

My new one is a 9700X. Didn't feel the need to spring for higher power budget for a marginal gaming performance bump. But I suppose that also means it's much more practical for me to jump to a newer CPU later.

Similarly, went from i7-4790K with 32GB RAM to 9800x3d with 96GB ECC RAM.

It's faster than the prior machine, but it sure does not feel like it does things the previous one didn't

Heh I also only sprung for 32 GB of RAM this time, which is still double my 4790K's 16 GB. But I don't use Chrome so RAM doesn't get used that heavily. ;)

I think it's very telling so many people upgrading now are coming from Haswell chips, they are a legendary chip generation, and arguably the last time anyone needed a CPU upgrade short of operating system support or warranty concerns.

>short of operating system support

Notably, Haswell makes the cut for Win11.

Not that I'd use that over Linux. (I run Arch, btw)

Officially it most certainly does not. Haswell chips are fourth-generation processors, Windows 11 requires eighth-generation processors. Though of course, if you bypass the checks it will install anyways.

I got confused.

It makes the cut for something else, x86-64v3.

Which I thought was the baseline for Win11. Apparently, not the case.

Had been running a coffee lake refresh 4 core for several years and as interested as I was in new platforms, especially AM5, the work of replacing motherboard never felt worth it. Now with the ram wars heating up, I just committed to that by picking up a used top-end 8 core coffee lake for $50 to cut a few seconds off my vulkan shader compiles with minimal replacement effort.

I really wish I would've bought 192G when it was less than a few thousand dollars!

Heh. It was a luxury purchase at the start of the year when I was only worried about tariffs. Wanted to lock in a new build good for years. Every once in a while I have a machine learning project that needs over 100GB and so it is nice not to have to overthink things. Honestly, I’m kicking myself I did not go all the way with 256GB.

I assume you're using 4 modules of memory, so the while the capacity is high, the bandwidth is low.

Low being relative.

6000 C28 at 256GB using 4x64 is not at all on the bandwidth high end these days, but it's way more than DDR4 could provide.

Most boards support 4x64 at 5600+ now, some go to 6400 with it if you tune voltages and terminations. .

Depends wildly on what you're doing.

I'm a gamer, often playing games that need a BEEFY CPU, like MS Flight Simulator. My upgrade from an i9-9900K to a Ryzen 9800X3D was noticeable.

You say that, but DDR6 will double the memory bandwidth over DDR5. This means modern systems will go beyond 200GB/s memory bandwidth just for the CPU alone.

Cute but that's not how it works.

The singaling is double the transfers but the bandwidth does not double.

We've seen what ... 5 ... itterations now and people still get this wrong.

The doubling of transfers comes at the cost of latency and processing overhead, always, until the new standard matures and might match the predecessor latency.

Did dual channel DDR5 double dual channel DDR4 bandwidth? Short answer ... nope, not even close, and DDR4 4000 remains viable and trades blows with DDR5 to this day in application performance despite "losing" in sulunthetics.

> DDR6 will double the memory bandwidth over DDR5

Considering PC desktops. DDR4 is 3200 MT/s max JEDEC. DDR5 is available on AMD since 3 years and is 5600. DDR6 specification is almost finished. It looks like DDR5 will double performance just right before new DDR6 DIMMs appear. Thus I'd expect DDR6 to double the bandwidth just as late when the new memory standard arrives.

> DDR5 is available on AMD since 3 years and is 5600

Strange, I bought 64GB DDR5 6400MHz last year and apparently my motherboard can handle up to 7200MHz (or more with overclocking).

AMD's CPUs don't support more than 5600 MT/s without overclocking; they're still using the same IO die from Zen 4, so their memory controller is pretty outdated. Zen 6 should introduce a new IO die with a better memory controller, but for now 6000 MT/s is the fastest reasonable memory overclock for AMD desktops.

Intel's desktop CPUs from last year support up to 5600 MT/s with regular DDR5 DIMMs, or 6400 MT/s for CUDIMMs. Speeds higher than this are achievable, but are overclocking.

If your memory modules are rated for 6400 MT/s, they are most likely advertising the speed when using an Intel XMP or AMD EXPO profile to overclock the memory (and the CPU's memory controller). The JEDEC standard profile likely is no faster than 5600 MT/s. It's also possible that you bought last year a kit of CUDIMMs rated for 6400 MT/s without overclocking, brand new to the market at that time, and of no help whatsoever with any CPU that isn't an Intel Arrow Lake.

Now that you say that, I checked in bios and looks like you're right. I have 4 sticks of "DDR5-4800-16GB @6400MHz". It was probably marketing speed, but works stable and memtest didn't find any errors.

Though clarified at the start about Desktop, but missed JEDEC applying, of course, generally for the whole post.

And? What real world impact will that have for people typing up an email and browsing the web?

It majes a huge difference for local AI models.

But they are still gonna fab the Zen 6 chips. So for people already with AM5 motherboards populated with RAM but rocking a Zen 4 CPU this could be a good time to upgrade that CPU with your existing setup. You passing this generation just means less competition for those CPUs which should make them even cheaper.

My understanding is they’re using the same process time for cpus and gpus so they may just be able to reallocate it for datacenter gpus. Sure they’re behind but some of the AI companies have already made deals with them as they just want compute, any compute. So I think the effect might be less than some hope for

and do what, buy now-hideously expensive DDR6?

A socket change does not correspond with a RAM change. Intel has had like 10 sockets in DDR4 including their newest socket.