With the release of Alder Lake less than a week away and the "Lakes" thread having turned into a nightmare to navigate I thought it might be a good time to start a discussion thread solely for Alder Lake.
Suitable motherboards for the 12400 will cost $100+ more than suitable motherboards for the 5600X?It only wins on price if you exclude motherboard costs. Once you factor in a suitable motherboard for the 12400 or a 5600X the price/perf/power is pretty much a wash. If you are intending on doing an upgrade later on then you might swing ADL because you can probably drop in a Raptor Lake CPU in the future OTOH if you only game you might go Zen3 and drop in a 5800X3D which is going to release well ahead of Raptor Lake.
It has a hard limit of a 76W PPT. You cannot get the 5600X to surpass that under stock operation.
From the same Anandtech:It's 88 W PPT. The formula is 1.35*TDP.
Compared to other processors, for peak power, we report the highest loaded value observed from any of our benchmark tests.
12400 review at Hardwareluxx, tons of tests and numbers :
Core i5-12400 im Test: Ohne E-Cores zur günstigen und effizienten Gaming-CPU - Hardwareluxx
Core i5-12400 im Test: Ohne E-Cores zur günstigen und effizienten Gaming-CPU.www.hardwareluxx.de
Specificaly the power comsumption numbers with Cinebench R23 :
Messungen: Leistungsaufnahme, Temperatur und Energieeffizienz - Seite 9 - Hardwareluxx
Core i5-12400 im Test: Ohne E-Cores zur günstigen und effizienten Gaming-CPU.www.hardwareluxx.de
Suitable motherboards for the 12400 will cost $100+ more than suitable motherboards for the 5600X?
https://www.scan.co.uk/shop/computer-hardware/motherboards-intel/3518/3519/3520/3521/3522/3523As of now they do in the UK. I am sure pricing will come down somewhat but today the cheapest ADL motherboard you can buy is £200.00.
https://www.scan.co.uk/shop/computer-hardware/motherboards-intel/3518/3519/3520/3521/3522/3523
I am also in the UK now and these listings took literally two seconds to find, and it doesn't include some other motherboard OEMs who haven't announced B660/H670 series boards (CES being still ongoing).
4% better perf/w than the 5600x. Gaming consumption looks good as well at roughly 50W.
The 12400 is 40% more efficient in the Igor's Lab gaming test (part 1). So, your conclusion is flawed.That being said it s right that in games it does fare very well power wise, there s no games power for the 5600X unfortunately, but this should be about the same TDP numbers.
The 12400 is 40% more efficient in the Igor's Lab gaming test (part 1). So, your conclusion is flawed.
Igor's lab was using PBO auto, which based on the numbers means it was using PBO, hence why I am personally dismissing them as crap.Not to drag this around in circles once again, but people are claiming that the Igor's Lab numbers themselves are flawed, so using them doesn't add anything.
Just compare a graph from them with one posted from AT in this very thread within roughly the last page of each other:
The Igor's ones are for a specific benchmark and the AT ones are just peak for any, but unless one is measuring from the wall while the other isn't, there's a pretty obvious discrepancy between the 5600X results.
Address your issues to igorslab maybe? I had nothing to do with making this chart!Watts is a current draw measure. For total energy used, you have to use Joules. Understanding science is even better than just reading.
Definitions:
A watt is a unit of power, named after engineer James Watt, which measures the rate of energy transfer. A watt, in electrical terms, is the rate at which electrical work is done when one ampere (A) of current flows through one volt (V).
joule, unit of work or energy in the International System of Units (SI); it is equal to the work done by a force of one newton acting through one metre. ... In electrical terms, the joule equals one watt-second—i.e., the energy released in one second by a current of one ampere through a resistance of one ohm.
The 12400 is 40% more efficient in the Igor's Lab gaming test (part 1). So, your conclusion is flawed.
Quoting Flawed Charts is a fallacy on itselfAddress your issues to igorslab maybe? I had nothing to do with making this chart!
Not only does one say average and the other say peak, but igors is a measurement over the complete time to render while the anandtech one is taking a single peak point.Just compare a graph from them with one posted from AT in this very thread within roughly the last page of each other:
But is it flawed? If this is the amount of watts needed to finish the job?Quoting Flawed Charts is a fallacy on itself
Not only does one say average and the other say peak, but igors is a measurement over the complete time to render while the anandtech one is taking a single peak point.
But is it flawed? If this is the amount of watts needed to finish the job?
Unrestricted 12900K without a Watt cap would also finish the job, faster and very inefficient. PBO is very similar to unrestrictedBut is it flawed? If this is the amount of watts needed to finish the job?
Address your issues to igorslab maybe? I had nothing to do with making this chart!
Also watt/sec is an equivalent of joule so taking the watt numbers and dividing by time to completion would be the same.
"A watt-second (symbol W s or W·s) is a derived unit of energy equivalent to the joule.[27] "
Or since the work done was the same for all CPUs you can just leave out the time measurement and division.
It's not a scientific unit of work anymore but you see how much energy each CPU needed to complete a fixed unit of work.
Of course, it doesn't matter in this case given both CPUs accomplish said task in basically identical times...You can't have an average be higher than the peak. If Igor's is measuring average power at 90W, that means it is peaking even higher.
Amount of watts to finish a job doesn't make any sense. @Markfw is correct. Igor is taking average watts over the course of the render job. In other words, he is taking samples over time of the watts being consumed by the CPU throughout the length of the job and then taking the average of those samples. If you want to talk about the amount of energy used to finish a job, the measurement would be in Joules.
How many time do I have to say this.... He set PBO to auto, and since we all (except you) know the numbers are too high, PBO was on at the time of the runs, thats why they are too high, and hence, why all the numbers are flawed.Not only does one say average and the other say peak, but igors is a measurement over the complete time to render while the anandtech one is taking a single peak point.
But is it flawed? If this is the amount of watts needed to finish the job?
Of course, it doesn't matter in this case given both CPUs accomplish said task in basically identical times...
Address your issues to igorslab maybe? I had nothing to do with making this chart!
Also watt/sec is an equivalent of joule so taking the watt numbers and dividing by time to completion would be the same.
Not only does one say average and the other say peak, but igors is a measurement over the complete time to render while the anandtech one is taking a single peak point.
But is it flawed? If this is the amount of watts needed to finish the job?
7-Zip decompression hm, why is it slower vs i5 11400F/12mb L3 Cache.
How can be slower, if i5 12400F has higher IPC+bigger L3 Cache 18mb.
7-Zip is clearly not the tool to use to measure IPC.