AMD Ryzen 7000 Series “Zen 4” Processors Expected to Launch in Q3 2022

[Endgame]

Seems to me you buy more for FOMO than for SOTA. And that's fine to each their own.

I tend to balance both in my purchasing decisions. It matters when I have the excess capital to buy, and what is out at that time as opposed to where I am currently.

Before my 3900x I was running a 7700k, before that a 2600k. As you can see I tend to wait a few generations before I look to upgrade. Of course in that same lifetime I'd upgraded video cards. The main reason I didn't upgrade cPU's was the socket didn't support the next gen really. Now I have that option and will probably to 5900x here before too long.

There is the FOMO call again, which is... strange regarding most of the purchases on the list? Lets look at the value aspect of the 1080ti, assuming I had landed a 3090 FE on launch day.

1080 Ti launched in March 2017 for 699, and the 3090 launched in September 2020 yielding a lifespan of 42 months. That is 16.65 / month for the card, not counting the sale price of my 980, nor the sale price of the 1080ti after replacement. Now, lets say instead of buying the 1080ti on launch day, I had waited until 6 months before the 2080ti launch to buy. At that point, I may have been able to find a small discount on the 1080ti - lets say I got it for 599 (Though I don't think the price ever dropped that low). The 2080ti released in September 2018, so the lifespan of the card is now 30 months, costing me 19.97 per month. Furthermore, the resale price of the 980 I had previously likely had dropped $100 from March 2017 to Sept 2018 costing me even more in the long run.

I don't see any advantage to waiting to purchase the card late in its life cycle vs immediately at launch, as you pay more per month for no gain in performance while also costing you resale value on the card its replacing. The big complication now is Crypto ruining card availability - I will admit to FOMOing the 3080ti step up, because even though its a bad finance decision (as laid out in the previous paragraph), god knows if the wait list for the 4k series will be even worse.

 

[MadMummy76]

I can’t remember the last time I had enough video card performance. Even the 3080ti won’t max out cyberpunk on my old Dell U3011 with ray tracing enabled (without backing down quality via DLSS) let alone my newer monitors . I’m just not interested in gaming at low settings or on a low resolution panel.

I've been perfectly fine playing cyberpunk on a 2080ti, dlss on quality is industingusable to me from native while playing, so upgrading just so I could play it on native would be having it for bragging rights. Besides when I played cyberpunk 3080ti didn't exist yet.

Even if I was still using the Dell as my primary monitor, since I run folding@home 24x7 when not gaming, I’d still get use out of the higher power video card without it being a FOMO choice.

If you can use the extra performance for productivity then it's not useless obviously. The question is do I wanna invest $2000+ for a process to finish in 25 minutes instead of 30? And the answer is always no. If it was 5 minutes instead of 30, that would be another story.

 

[Brian_B]

I think a lot of the current Raytracing hype is FOMO - I don't feel like I'm missing out on anything there.

 

[LazyGamer]

I'm curious about the APU's (cue @Dan_D head exploding)

They should have all been APUs from the start - only now is AMD really understanding the importance of that, with Intel having had done it for over a decade now. Everything from lightweight games to hardware transcoding acceleration to just having a few more video outputs helps.

It's the same as mentality as "future" proofing a build. No such thing.

Taken to its logical extreme, absolutely not, but there's something to be said about making reasonable allowances for growth. A little more RAM, a little more VRAM, a few more cores, a few more PCIe lanes... that can hold off a larger upgrade down the road if need be.

 

[LeRoy_Blanchard]

Taken to its logical extreme, absolutely not, but there's something to be said about making reasonable allowances for growth. A little more RAM, a little more VRAM, a few more cores, a few more PCIe lanes... that can hold off a larger upgrade down the road if need be.

To an extent. If you can't benefit from those "a little more" things now, then saving the extra money it costs now and putting it towards a new build sooner would be a better option.

 

[LazyGamer]

To an extent. If you can't benefit from those "a little more" things now, then saving the extra money it costs now and putting it towards a new build sooner would be a better option.

Agreed, there's no way to know if it would be worthwhile or not.

 

[Zarathustra]

Taken to its logical extreme, absolutely not, but there's something to be said about making reasonable allowances for growth. A little more RAM, a little more VRAM, a few more cores, a few more PCIe lanes... that can hold off a larger upgrade down the road if need be.

I generally consider future-proofing to be a fools errand.

Tech generally drops in price rather quickly (current crazy market excluded)

Unless you are in an Apple type of situation where - for instance - the RAM is soldered to the motherboard - buying more now generally costs more than upgrading later when you need it, and winds up being a waste.

I know some people treat their PC' like they are some sort of static build. They build it once, then sell it or decommission it, and build another one. I guess if that's what you do, then maybe it makes more sense, but that "static build" mentality never made any sense to me.

To me my PC is the spiritual successor of my first 286 I got back in 1991, upgraded one component at a time. There may not be any part left from the original one, but I never just threw everything out at started from scratch. In most cases upgrades were one little piece at a time that gradually changed the nature of the machine. In others, lots of parts may have been changed at the same time, but there was always a few parts that were kept.

I'm typically in there tinkering and replacing parts every few months, upgrading or replacing one thing or another. The PC never gets to stay in one configuration for very long.

So, if I am going to decide between spending lots of money now buying more than I need, or upgrading later when I actually need it for less money, that's a very easy choice to make. I always upgrade later.

 

[LazyGamer]

It's always a coin flip to be sure. I guess another way to put it would be to make sure that you get something that is upgradeable, and if you can't, then make sure you don't undershoot.

All of my systems / components either get put to server use or handed on, generally speaking.

 

[Endgame]

Ram is one thing that I don’t mind buying more of, unless I’m at the start of a new generation of Ram. After about a year, the price doesn’t drop dramatically, and if it starts to get “too” low, there just happens to be an earthquake, or fire, or flood, or whatever to prop the price back up. I’ve also not had the best luck with timing of mixed size kits, especially mixing manufacturers. In my most recent system, I needed about 40gb Ram. I could have bought a 32gb kit and an 8gb kit, but I spent like $100 more to get a 64gb kit.