I have a laptop (i7-7700HQ CPU @2.80GHz with 16GB RAM) that can handle tracking, and fairly complex processing with track freezing, etc. However, I want to work with advanced mix templates for improved workflow. My dilemma is should I get a 5900X Zen 3 based system now, or wait until next year for a faster and more efficient Zen 4 5nm CPU? Anybody else having similar thoughts?
If you can make do with your current setup, wait. The could be a Zen 3 refresh this Fall/Early Winter (Zen3D?). Barring any further market disruptions Spring 2022 could offer some excellent choices from AMD, Apple & Intel. If you build/purchase now, go with the 5800X or 5950X and be sure to use memory tested & validated for the motherboard that it will reside in. Recommend: G-Skill Trident Z Neo
I have a 5900x and cannot fathom anyone needing anything faster. It gives you freedom.. you are no longer tied to your cpu. you forget about your cpu and can do anything you want.
I agree with you totally, however I have a very bad Nebula habit that needs a lot of processing power
It (simply) depends on how much $$$ would having a fast machine get you in a year 5900x here (at home, for a/v), same ^^ same, again ^^ A good mobo+RAM (dark hero + 64GB crucial ballistix here) gives a super stable system (and I'm lazy at tweaking it endlessly)
Not sure if this helps but it compares a 2 year old i9 with a 5950x Looks like he got twice the performance over the i9
You would need to do truly *Massive projects of Epic proportions to really max out any of the current AMD cpu's.. As long as it's got 8cores/16threads or more, you're good to go.. Waiting for the ever next gen, what you win is less Power Consumption/Heat, and that's positive too. Or perhaps in this case this new 3D cache could improve raw core power, and maybe latency, who knows.. *(Ofc one could Max it out being silly like, putting EQuilibrium at max-res linear-phase on each of your 30 instrument buses.. Or melding togheter the Composing/Production session with 100 Kontakt instances and VSTs, with the Mastering session with all the FX in the planet, etc.. If well done, I guess one could save time with that, but it's kind of a silly way to abuse the computer.. With these new Cpu's one could do Anything with Audio/Music, and any shortcoming would be just a matter of managing it properly/more efficiently)
Yes, Intel are attempting a comeback too, so it could also be interesting. I can easily continue with my current workflow even though it requires a lot of discipline, but that's not necessarily a bad thing since it makes you focus on attention to detail as you consider every move. It has definitely helped to finetune my analytical listening skills. I always start with the drums a bass, and then work on the instrument groups and vocals, while committing the edits as I go. I think working with restrictions is a good way to learn so I have no regrets, but it is rather cumbersome and I look forward to freeing myself from all of these restrictions! Last edited: Aug 16, 2021
Have you tried Nebula yet? It's a really painful process and it consumes a lot of resources, sometimes 5 instances of the plugin just for one EQ, plus multiple instances for compression, consoles, tape, reverb, group & master bus treatment, etc. The Nebula crowd have been waiting 10+ years for this kind of computing power Last edited: Aug 16, 2021
Dude... I made music with Pentium I 200 overclocked to 233 MHz. Admittedly using only outboard hardware, but anyway... I used a lot of plugins. Don't ever wait for anything. Make music with whatever you've got.
Hmm not really.. I began trying it at some point, but I found it unnecessarily complex for what I'm doing.. Also it triggers my OCD a bit to think of all the variables that go behind the modules, like, this person measuring that component/chain of components and dialing it into the platform, etc.. I donno how accurate/reliable it can be, or how much of an improvement it could represent. But otoh I know regular top-tier VST are made to a great level of Perfection, and they do what I need pretty much flawlessly. Also since I work on Videogame Soundtracks, I'm not really too much into Analog emulation per-se, I prefer to do good timbre selection/adjustment, and keep things as Clean/clear, balanced and correct as possible on their own.. So I use plugs with a clean philosophy in mind like: Pro-Q3, Oxford EQ on the Intruments, EQuilibrium on the Master.. Pro-C2 (oversampled)/TB BusComp on the Instruments, Elysia Comp and MB-5 on the Master.. And ofc a bunch of algo/convo Reverbs; Ozone and other little things here and there. But yeah the most demanding is EQuilibrium by far.. It's only from time to time that I find some element might need more Mojo, or that I'm requested to record real Guitar, for which I could try going for a more analog-like chain of fx.. So yeah, I did look into it and tried it a while ago, but since I've got everything pretty well laid out/solved, I haven't had the chance to really dig into it. -On the topic of Analog-ness tho, I've been trying this non-linear summing console plugs like Waves NLS, Brainworx TMT etc.. I find the idea very Interesting, and I found I like more the Waves Mike/API model over the SSL or Neve.. And that I prefer Brainworx take on SSL over Waves Spike, (although I haven't tried the 9000J yet) Besides, I haven't had the chance to try it, but judging by the Demos, the one that impressed me the most is UAD Luna Summing thing.. So yeah I like the concept, and I can see/hear some of the benefits/realism added. But for similar reasons I didn't really end using it.. for the type of music I'm working on it's not so necessary, and while it adds a certain Realism, there's a part of it that doesn't really go with me, because there's necessarily gonna be certain "imperfections" that affect the balance of freqs/phase/volume etc, which would be out of my reach/hard to evaluate. Ofc I'm not like super OCD, but If I worked for days/weeks on Balancing something trying to get a certain level of "perfection", I'm not gonna randomly throw a bunch of "unbalanced" channels on it, just because it gets some extra sheen/subtle depth.. If an element needs more sheen/depth I can work it out, but "staining" everything with that feels a bit radical to me atm. However who knows, maybe someday I can put myself to record/produce more real/Guitar based music, then it could be really handy.. Last edited: Aug 16, 2021
Good post I'm actually looking for the coloration found on classic recordings, so it's purely a case of aesthetics for me.