• he/him

I occasionally write long posts but you should assume I'm talking out of my ass until proved otherwise. I do like writing shit sometimes.  

 

50/50 chance of suit pictures end up here or on the Art Directory account. Good luck.

 

Be 18+ or be gone you kids act fuckin' weird.

 

pfp by wackyanimal


 

I tag all of my posts complaining about stuff #complaining, feel free to muffle that if you'd like a more positive cohost experience.

 


 
Art and suit stuff: @PlumPanAD

 


 
"DMs":
Feel free to message as long as you have something to talk about!


Short answer: Bad, but a symptom moreso of the rapid rise in power draw ahead of cooling, rather than there just being TOO much power use for any situation. A 3.8GHz P4 consumes about as much power as a 5800 X3D does.


Long answer: well,

First of all, at least on the early 100W+ P4s (early stepping Prescotts), the TJMax was 75C. If you hit 75C, the CPU started throttling. It's really, REALLY hard to keep a 100W chip under 75C without a lot of noise, even today.

Yes we had slightly better coolers available back in the day, but the point here was that Intel was shipping a chip stock that could not be cooled by the cooler that was sold with it. But, the "rule of thumb" at the time was to try and keep your CPU at or below 60C if you could. Those were the bad old noisy days.

I had thought that the act of checking the actual power going to the CPU via 12V rails, rather than total system power draw, didn't come around until the 2010s. But Tom's did in fact do this as part of their testing of the E0 stepping 3.8GHz chip. The Tom's site is so bad now that I'm just going to screencap the relevant part.

With the clamp they measured around 105W real power use. This is about the same as what a 5800X3D will use, give or take a few watts, as a modern reference. Obviously there's a difference between a modern 8 core extended cache chip compared to a single core, dual thread chip from 20 years ago, but again this is just to give a point of reference for how much total power the CPU is using.

Between then and now, heatpipe coolers became extremely common. CPU coolers are WAY bigger too. PWM fans are expected everywhere, not just on the CPU fan. Fans are bigger, both for more airflow and less noise. Cases have WAY more airflow, both from better chassis designs and also from there being a lot less crap in the way (drive bays). CPUs have tons of different power modes, and the better ones actively adjust their clock speed based on temperature and power draw. These early P4s had two modes: On, and throttle. Soon after they started getting "lower the clocks at idle". CPUs are also built to run WAY hotter, safely. That's all why 105W of power draw was a Big Fucking Deal back then, whereas now it's a very efficient chip if anything.

I think modern desktop CPUs drawing 200w+ are fucking stupid. If you have some gigantic server chip doing workstation things that draws 200w... sure, whatever. I still don't like the power draw but I see that as having more of a purpose, plus a lot more silicon to keep lit up. Desktop parts just get shoved WAY up the volt/freq curve for marketing reasons and I only find that upsetting.


You must log in to comment.