AMD's Most Efficient CPU: Ryzen 7 5800 Non-X Review, Benchmarks, & Power
video description
Date: 2021-08-01
Related videos
Comments and reviews: 10
Ted
I notice that when comparing the non-X to the X varient you always seem to bring the non-x upto the x varients freq and such to get a closer comparison if you can. With the non-x being more efficient and in some benchmarks being within spitting distance of the x varients performance I am wondering if it isn't possible to work the other way around. I mean, would you not get a better idea of just how much more powerful/efficient, in terms of performance per watt the non-x is by trying to get the x varient to perform at the same power as the non-x varient and then see if the X can do the same as the non-x at or close to its powerlevel. Would this not give a better understanding than just trying to strangle as much performance out of the non-x by ramming it full of power and clocking it up?
reply
I notice that when comparing the non-X to the X varient you always seem to bring the non-x upto the x varients freq and such to get a closer comparison if you can. With the non-x being more efficient and in some benchmarks being within spitting distance of the x varients performance I am wondering if it isn't possible to work the other way around. I mean, would you not get a better idea of just how much more powerful/efficient, in terms of performance per watt the non-x is by trying to get the x varient to perform at the same power as the non-x varient and then see if the X can do the same as the non-x at or close to its powerlevel. Would this not give a better understanding than just trying to strangle as much performance out of the non-x by ramming it full of power and clocking it up?
reply
EUROSCI
If you are interested in efficiency, why don't you report the total energy consumed to carry out a certain job, for instance the total joules consumed to render your logo or any of the other tests you made? You currently talk about efficiency in the title but only provide data about power draw and time to do a certain job, but you do not present the total energy consumed. You could say it would be possible to calculate the energy efficiency of any of your tests by multiplying the power draw by the time to do the job, but besides being very inconvenient, this would only work if you provided average power draw, and I'm not sure that's the case. Currently your review seems to be either unfinished or not delivering on its title's promise.
reply
If you are interested in efficiency, why don't you report the total energy consumed to carry out a certain job, for instance the total joules consumed to render your logo or any of the other tests you made? You currently talk about efficiency in the title but only provide data about power draw and time to do a certain job, but you do not present the total energy consumed. You could say it would be possible to calculate the energy efficiency of any of your tests by multiplying the power draw by the time to do the job, but besides being very inconvenient, this would only work if you provided average power draw, and I'm not sure that's the case. Currently your review seems to be either unfinished or not delivering on its title's promise.
reply
Bryan
Doesn't matter, 2 thirds of all online gamers are hackers more hackers then non hackers I uninstalled all online games anything with a credit card attached and selling my pc I'm done... no need to be the fastest when the laptop with a hack will always win at just one click. all garbage photoshop, rendering movies , business that's it everything else some sort of something in the software killing it so no need to be a programmer for hardware when you realize 2 thirds of your time will be to stop dumb people from cheating. then you realize 2 thirds of the planets population is dumb as a bag of rocks.. to advanced .. to soon. see you in another 15,000 years or so.
reply
Doesn't matter, 2 thirds of all online gamers are hackers more hackers then non hackers I uninstalled all online games anything with a credit card attached and selling my pc I'm done... no need to be the fastest when the laptop with a hack will always win at just one click. all garbage photoshop, rendering movies , business that's it everything else some sort of something in the software killing it so no need to be a programmer for hardware when you realize 2 thirds of your time will be to stop dumb people from cheating. then you realize 2 thirds of the planets population is dumb as a bag of rocks.. to advanced .. to soon. see you in another 15,000 years or so.
reply
Speak
I do wonder if AMD will ever properly supply chips like this, the 5700G, etc to the DIY market? Is it just the current production limitations that are stopping them or is it internal decisions?
Strangely, if you really want to buy these AMD chips or/and would like AMD to reduce their prices then you have to pray that Intel properly comes out swinging with Alder Lake with a full line up with amazing performance and at great pricing in an attempt to win hearts and minds back from the last 3 years where almost the only decision that made sense was an AMD CPU. Sadly I don't think Intel will do any of that but I would love to be wrong.
reply
I do wonder if AMD will ever properly supply chips like this, the 5700G, etc to the DIY market? Is it just the current production limitations that are stopping them or is it internal decisions?
Strangely, if you really want to buy these AMD chips or/and would like AMD to reduce their prices then you have to pray that Intel properly comes out swinging with Alder Lake with a full line up with amazing performance and at great pricing in an attempt to win hearts and minds back from the last 3 years where almost the only decision that made sense was an AMD CPU. Sadly I don't think Intel will do any of that but I would love to be wrong.
reply
Dallas
I limit my 5800X to 75 watts because it honestly doesnt need more than 70. Headroom or leeway is always good and provides a good buffer for limiting. I also limit it to 1.27V as well, and will boost longer and higher (5.1 ghz) like this in games. With an actual performance boost to boot. Lower temps, ect. win/win/win. Downside is max cpu load performance takes a small hit... %6? maybe
Sould also be noted that most bios are pretty poorly configured for consumer ryzen 5000 chips. Usually too low vcore and too high coreV
reply
I limit my 5800X to 75 watts because it honestly doesnt need more than 70. Headroom or leeway is always good and provides a good buffer for limiting. I also limit it to 1.27V as well, and will boost longer and higher (5.1 ghz) like this in games. With an actual performance boost to boot. Lower temps, ect. win/win/win. Downside is max cpu load performance takes a small hit... %6? maybe
Sould also be noted that most bios are pretty poorly configured for consumer ryzen 5000 chips. Usually too low vcore and too high coreV
reply
Zonker
I had an Alienware laptop around 2005. It was a tank and got as hot as the sun. I had it sitting on a cooling rack thing... sent it back twice, they never did a thing. Went to the UK on a trip and was told Don't worry about the warranty expiring while you're away, we'll honor it . Came back... you're warranty has expired... click My question is: How is this company still in business? They must have the greatest PR ever. I know this video was on a Ryzen 65w CPU... but I saw the teardown of that Alienware pig again.
reply
I had an Alienware laptop around 2005. It was a tank and got as hot as the sun. I had it sitting on a cooling rack thing... sent it back twice, they never did a thing. Went to the UK on a trip and was told Don't worry about the warranty expiring while you're away, we'll honor it . Came back... you're warranty has expired... click My question is: How is this company still in business? They must have the greatest PR ever. I know this video was on a Ryzen 65w CPU... but I saw the teardown of that Alienware pig again.
reply
Smokey
After learning everything I have over the last month or so even though it s only a few FPS better I m gonna go with a Ryzen 9 5900X, where I could save some money and get the 7 5800X but games are becoming more dependent on more cores 12 cores 24 threads is plenty, the crazy thing is the 5900X is way better than an AMD 2nd Gen Ryzen Threadripper 2950X which is a lot more expensive and kinda crazy it s 16 core 32 threads but the 5900X even the 5800X beats it
reply
After learning everything I have over the last month or so even though it s only a few FPS better I m gonna go with a Ryzen 9 5900X, where I could save some money and get the 7 5800X but games are becoming more dependent on more cores 12 cores 24 threads is plenty, the crazy thing is the 5900X is way better than an AMD 2nd Gen Ryzen Threadripper 2950X which is a lot more expensive and kinda crazy it s 16 core 32 threads but the 5900X even the 5800X beats it
reply
mookieliciousness
I picked up a 3900 none X as part of a PC upgrade, 12 cores with / 24 threads at 65W TDP...running it with a 212 Evo setup for push pull, maxing out temps at 67C under load in gaming and production use. I upgraded from a 2600K so it was an AWESOME jump in spec. I had a choice between a 5600X and the 3900 none X, I opted for more cores as I game and work from the same rig, no regrets.
reply
I picked up a 3900 none X as part of a PC upgrade, 12 cores with / 24 threads at 65W TDP...running it with a 212 Evo setup for push pull, maxing out temps at 67C under load in gaming and production use. I upgraded from a 2600K so it was an AWESOME jump in spec. I had a choice between a 5600X and the 3900 none X, I opted for more cores as I game and work from the same rig, no regrets.
reply
BRUXXUS
That efficiency is truly bonkers! The fact that the 5800 is likely lower quality silicon than 5800X makes me wonder how efficient you could make a 5800X with tweaks.
Now that I think of it... that would be a fun new sort of... overcloc... er.... competitive computing? Tweaking things to see what max performance you can get from least power.
reply
That efficiency is truly bonkers! The fact that the 5800 is likely lower quality silicon than 5800X makes me wonder how efficient you could make a 5800X with tweaks.
Now that I think of it... that would be a fun new sort of... overcloc... er.... competitive computing? Tweaking things to see what max performance you can get from least power.
reply
Vlad
Sooo, it seems like amd is running these X parts at a completely non-efficient range just to play in the same league as Intel, the Ghz race. TBF, intel is also paying the same costs to go over 5Ghz.
I would like to see a reset on this like for example core 2 duo, which had a 1.83Ghz E6300 cpu that beat a pentium 4 at close to 4ghz.
reply
Sooo, it seems like amd is running these X parts at a completely non-efficient range just to play in the same league as Intel, the Ghz race. TBF, intel is also paying the same costs to go over 5Ghz.
I would like to see a reset on this like for example core 2 duo, which had a 1.83Ghz E6300 cpu that beat a pentium 4 at close to 4ghz.
reply
Add a review, comment
Other channel videos
