Remember when some of ya'll said that it was too early to decide whether AMD was "back"?

How about now:

PS: Not an "in your face", it's a serious question

Oh heck, the productivity test poor things! I know where I'm putting my money in my next build.

Ryzen 7 3700X @ $329 is closer to the i9 9900K @ $499

And also consider that the i9 only has 16MB L3 cache while the Ryzen 7 has 36MB.

Like, it's clear when you want better bang for your buck, you'd be going AMD.

@brandon not silly response:

Oh wow that seems like a big step toward making not-just-fan-based-cooling more of a default in desktops.

@emsenn the on-mobo cooling seems to be fan-based though, did I miss what you intended to say?

@brandon Oh, no, I misassumed and thought it meant like, it needed a place to hook into liquid cooling from the mobo

@emsenn Oh no, there will literally be fans ON the mobo cooling the mobo chipset

@brandon interesting!

I wonder what the co2 output of using a computer with all this for text editing for an hour is, versus something like an RPi.

(I know this CAN do much more, I'm curious about its footprint when it's doing nominal work.)

@emsenn co2 output? I don't think that computers exhaust co2 🤔

@brandon Sorry, what I mean more directly is I wonder what their electrical use is, and what that means given the average power plant's co2/wh rate.

@brandon For example, sitting "idle" with me using it as a text editor, my desktop+monitor pull about 80 watts, while my laptop pulled about 11 watts.

That's, if I recall, about a difference of ~0.02g an hour, or ~0.16g over a day, or ~58g a year.

Which isn't much, but that's still 58g of wasted energy, just because I'm typing into one computer instead of another.

@emsenn I think the numbers will change depending on the source of the electricity. For example hydroelectric vs. burning coal

@brandon Which is why i said based on the average power plant output? Come on man, we're talking averages, you can't go "but specifics." Logic fallacy yellow card.

"Me doing the thing is okay" doesn't change how many people get their energy from non-renewables, that you're moving the overton window of consumer needs toward higher power-consumption, etc.

@emsenn I really don't want to participate in this discussion.

@brandon Don't get defensive just because I posed a question about your energy use. I wonder what the difference is - you went from "that's a nonsense question" to "that's a nonsense question" but no it's really straight-forward: what's the difference in average co2 output to run the computer at idle.

@brandon I actually didn't even ask about YOUR energy use, you alluded to that. I just asked about the machine's use. (Sorry, I'm done now.)

@brandon (And honestly, I use my computer much more than 8h a day.)

Now acknowledge that multiple people in the world are using computers that draw more power than they need to idle: think about how many workstations you handle for your job, you listed it on your profile recently, it was legit dozens.


Assumed numbers:

RPi 4 Max = 6W
Ryzen 7 Idle = 70W (including other parts of computer)

Calculated per hour co2 output:
Vermont: 0.003g co2/hour
Wyoming: 0.925g co2/hour

In vermont:

RPi4 = 0.018g/h co2 output
Ryzen = 0.21g/h
Delta = 0.192g/h

In Wyoming:

Rpi4 = 5.55g/h
Ryzen = 64.75g/h
Delta = 59.2g/h

You can see why I said "it depends" :P

Source for co2 output numbers used in calculations:

Conclusion: Rpi4 produces 8% of the co2 the ryzen 7 PC might

@brandon I think you're overcomplicating it by considering power plant differences yourself instead of working with a known average. The question is "what is the difference in co2 output between the ryzen and Rpi." The math can be simple, (x-y)*(z) where x is the kw/h of the ryzen, y the kw/h of the RPi, and z our average co2/kw*:

(0.07-0.006)*512 = 32.78g/kwh difference in output.

* there's debate about what "our" average is, I went for the U.S.

@brandon I don't mean to be obtuse but your approach just didn't answer the question I posed, which was to do with the difference in co2, not the proportions/ratios of co2.

@emsenn Did you mean to say 32.78g/kwh or per watt-hour?

@brandon I believe per kwh but I might've misreckoned, arithmetic is not my strong suit.

@emsenn ugh fuck it :P We know the difference is "a lot" :P

@emsenn If you would have said watt-hours (which seems to make more sense to me considering my calculations) I would have said the average of the two deltas results in 29.7g/h which would have been pretty close to your 32.78

@brandon I don't /think/ I misreckoned, for what it's worth.

While it's a lot in proportion, 38g/kwh works out to an extra 2.2g/h for the ryzen.

According to this and some math, the average co2 output for a person is 569g/h, so a ryzen is ~0.03% of that.

Which I'm hesitant to draw any conclusion based on until I were to measure what % of the daily average other differences in choice could make.

@brandon It really is some bullshit how much just... slogging through basic math it takes to be reasonable; the results for these questions should honestly be easily searchable online.

No, my brain has way more cache and it doesn't cost me a buck.

@brandon I’m pumped. Hoping to get some kit at work to play with.

@boblmartens o.o 🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤🤤

@brandon I need to redo my standard 1U buildout again. 8 months since the order is a long time.

@boblmartens What do the servers do in this context? I know they're virtualization hosts but, I guess I'm asking what your organization does

@brandon We are a college so mainly file servers, application servers, database servers, web servers, etc. We are running XCP-ng 7.6 right now and I am looking forward to the updates kernel in the 8.0 release.

@brandon I’m hoping we can look at some Lenovo ThinkPads as replacements for faculty machines.

@boblmartens oof, not much of a choice when it comes to AMD thinkpads 🤔

@boblmartens I think there was a time when there was NO AMD thinkpads, right?

@brandon Yes. First purchase may be a replacement for my venerable X220 I use for sysadmin duties.

@boblmartens Which ThinkPad(s) are you looking at to replace it?

@brandon A285 and X395 right now. I may go larger as well, need to sit down and do some thinking.

@brandon Looking over things, the T495 looks pretty amazing ... tempted ...

@brandon not at all if making a hackintosh, then you need an extra graphic card

@brandon hackintosh has some very specific requirements to work. one of them is a graphic card - it only works with Intel's integrated graphics and some AMD Radeons (no integrated ones). if I remember well, it's about Metal rendering support. anyway, this means that if you want to use AMD CPU in a hacc, you also need an extra graphic card.

@rnickson so if you were going to get a graphics card anyway, still AMD the better case?

@brandon it needs some extra effort from you to set up (modified kernel needed - look at ), but looking at the price difference, probably worth it

@brandon ✴️ Note to self: If I ever decide to get a desktop, consult with Linux Liaison first!

> On newer Linux distributions, there's a hard regression either within the kernel but more likely some cross-kernel/user-space interaction issue leaving newer Linux distributions unbootable.

But benchmarks looks promising.

@ashwinvis Luckily I'm looking more to purchase next year :D

Sign in to participate in the conversation

Fosstodon is a Mastodon instance that is open to anyone who is interested in technology; particularly free & open source software.