Share your computer setup!

[AVLNCH]

Do you have a speedy new system that you’d love to show off to the world? Perhaps you’re someone who just plays casually on the home PC? Why not post your setup in here to share & compare with other members of the Steam community!

[N4K]

NVIDIA GeForce RTX 4070 Ti

Procesor Intel Core i9-13900K

Intel Core i9-13900K Processor
NVIDIA GeForce RTX 4070 Ti
ASUSTeK COMPUTER INC. ROG STRIX Z790-H GAMING WIFI


Memory65.536 MBModule 132.768 MB Kingston @ 5.600 MHzModule 232.768 MB Kingston @ 5.600 MHz

[xSOSxHawkens]

Publicado originalmente por ChickenBalls:

its not a 3090/3090Ti

https://www.techpowerup.com/img/5wTpTV9WTdU7ZlZe.jpg

In the chart above, NVIDIA shows how current spikes get mitigated by their new VRM design, which uses a PID controller feedback loop. While the RTX 3090 bounces up and down, the RTX 4090 stays relatively stable, and just follows the general trend of the current loading pattern thanks to a 10x improvement in power management response time. It's also worth pointing out that the peak power current spike on the RTX 3090 is higher than the RTX 4090, even though the RTX 3090 is rated 100 W lower than the RTX 4090 (350 W vs 450 W).

please show a video of someone with 5800X3D/7800X3D and a good 850W PSU experience shutdowns because of transient spikes.

Dude... lmao...

You are on the wrong side of facks chickenballs, chill. Transients being able to down quality 850's have been a thing since the Vega line of GPU's on AMD and since the RTX3k line on Nvidia.

Flagships have needed 1kw PSU's for awhile now for real stability. Not saying you cant use less, but there will, inevitably, be eventual crashes (which will probably get blamed on something not at actual issue).

[Jamebonds1]

Publicado originalmente por ChickenBalls:

1000W is only required if you have an inefficient CPU like 13900K
850W is more than enough for 5800X3D/7800X3D and 4090

There is no inefficient on Intel 13th Gen except PSU.

Keep in mind, Plus 80 mean nothing unless you read the reviews where PSUs was deeper reviewed. Plus 80 only proof efficient by power factor, nothing else for stability verification.

[ChickenBalls]

are you saying that gtx 1080ti and 2080ti required 1000w psus?

and I can provide credible sources for my claims
https://www.techpowerup.com/299096/nvidia-details-geforce-rtx-4090-founders-edition-cooler-pcb-design-new-power-spike-management?cp=2

[MonkehMaster]

Publicado originalmente por ChickenBalls:

its not a 3090/3090Ti

https://www.techpowerup.com/img/5wTpTV9WTdU7ZlZe.jpg

In the chart above, NVIDIA shows how current spikes get mitigated by their new VRM design, which uses a PID controller feedback loop. While the RTX 3090 bounces up and down, the RTX 4090 stays relatively stable, and just follows the general trend of the current loading pattern thanks to a 10x improvement in power management response time. It's also worth pointing out that the peak power current spike on the RTX 3090 is higher than the RTX 4090, even though the RTX 3090 is rated 100 W lower than the RTX 4090 (350 W vs 450 W).

please show a video of someone with 5800X3D/7800X3D and a good 850W PSU experience shutdowns because of transient spikes.

google?

also, im not going to run over and reinstall an 850w in the persons pc i was talking about earlier, just to not only hassle him, but make a nonsense video showing his pc crashing while playing games with a 850w and said crashing vanish from existence after replacing it with a 1000w , but also you might ask, why i chose this solution, well.. because its common information across the internet that crashing/shutdowns happen when a 4090 power spikes because of under powered psu's being used with them and the 850w being the common choice, through bad recommendations.

heck even newegg and its sellers are recommending 1000w psus for 4090's. well at least the ones i skimmed through earlier today.

anywho, not worth wasting time arguing over, google your proof.

Publicado originalmente por ChickenBalls:

are you saying that gtx 1080ti and 2080ti required 1000w psus?

and I can provide credible sources for my claims
https://www.techpowerup.com/299096/nvidia-details-geforce-rtx-4090-founders-edition-cooler-pcb-design-new-power-spike-management?cp=2

what does 1080's or 2080's have to do with a 4090?

[ChickenBalls]

Publicado originalmente por Jamebonds1:

Publicado originalmente por ChickenBalls:

1000W is only required if you have an inefficient CPU like 13900K
850W is more than enough for 5800X3D/7800X3D and 4090

There is no inefficient on Intel 13th Gen except PSU.

Keep in mind, Plus 80 mean nothing unless you read the reviews where PSUs was deeper reviewed. Plus 80 only proof efficient by power factor, nothing else for stability verification.

https://www.techpowerup.com/review/intel-core-i9-13900k/22.html

167W average in games at stock speed
287W in multi-threaded workloads
413W OC all core

nobody was talking about PSU ratings

[Jamebonds1]

Publicado originalmente por ChickenBalls:

Publicado originalmente por Jamebonds1:

There is no inefficient on Intel 13th Gen except PSU.

Keep in mind, Plus 80 mean nothing unless you read the reviews where PSUs was deeper reviewed. Plus 80 only proof efficient by power factor, nothing else for stability verification.

https://www.techpowerup.com/review/intel-core-i9-13900k/22.html

167W average in games at stock speed
287W in multi-threaded workloads
413W OC all core

nobody was talking about PSU ratings

Have your read what was in the table?
This was a program running not cpu itself. As I said, CPU can't be inefficient because it is a power loading based on activities. Same true for power supply with load resistor.

[ChickenBalls]

Publicado originalmente por Jamebonds1:

Publicado originalmente por ChickenBalls:

https://www.techpowerup.com/review/intel-core-i9-13900k/22.html

167W average in games at stock speed
287W in multi-threaded workloads
413W OC all core

nobody was talking about PSU ratings

Have your read what was in the table?
This was a program running not cpu itself. As I said, CPU can't be inefficient because it is a power loading based on activities. Same true for power supply with resistor load.

uh yes....

https://tpucdn.com/review/intel-core-i9-13900k/images/efficiency-gaming.png

1.50 frames per Watt at 1080p vs 4.1 frames per Watt for 5800X3D

im not the saying its inefficient here
if you have problem with that statement tell all the review sites like guru3d, Techpowerup or Anandtech

[xSOSxHawkens]

Publicado originalmente por ChickenBalls:

are you saying that gtx 1080ti and 2080ti required 1000w psus?

and I can provide credible sources for my claims
https://www.techpowerup.com/299096/nvidia-details-geforce-rtx-4090-founders-edition-cooler-pcb-design-new-power-spike-management?cp=2

You dont read do you...

Publicado originalmente por xSOSxHawkens:

...since the Vega line of GPU's on AMD and since the RTX3k line on Nvidia.

Do you see GTX (anything) or RTX2k in the above?...

You then linked to pre-release Press Release piece about all the Talking Nvidia was doing about their tech before anyone had their hands on it.

I will at least give you that techpowerup is a good source *but* not when they are parroting a pre-release PR piece, and also *not* when they as a pulication/company litterally lack the hardware needed to even test the claims they are parroting from NV.

To be clear, the 4090 is better about spikes than the 3090. But its still not great. Also to be clear spikes do *not* equate to normal/average power use, nor are you talking about the types of spikes you, me, or most anyone else can see.

In order to even be able to log/see these types of spikes you *have* to use scientific equipment, often to lab grade standards. There are only a handfull of review/publication sites world wide who are setup to actually test this stuff.

https://www.youtube.com/watch?v=j9vC9NBL8zo&t=615s

Watch here (&t=615s if it didnt time stamp right) and you can see the testing needed. You can also see that though its not as bad as people woried, the 4090 is not great either.

Point is, they days of 1KW psu's are here again.

https://www.digitaltrends.com/computing/the-decade-of-ignoring-your-gaming-pcs-power-supply-is-over/

[ChickenBalls]

Flagships have needed 1kw PSU's for awhile now for real stability.

hmm I wonder who said that




https://www.tomshardware.com/news/1200w-power-requirement-rtx-4090

Recommended wattages should be taken with a grain of salt, since they won't apply to everyone's personal rig. Some might have a Core i9-12900K overclocked to 5.5GHz that consumes over 300W of power at its peak; others might be running a Ryzen 7 5800X3D that consumes less than 100W of power while gaming. Meanwhile, a Core i9-10980XE with a full blown RAID 10 setup and 10 hard drives could consume gobs of power. All of these combinations result in different levels of PSU power, regardless of what the AIB partner suggests.

[Jamebonds1]

Publicado originalmente por ChickenBalls:

Publicado originalmente por Jamebonds1:

Have your read what was in the table?
This was a program running not cpu itself. As I said, CPU can't be inefficient because it is a power loading based on activities. Same true for power supply with resistor load.

uh yes....

https://tpucdn.com/review/intel-core-i9-13900k/images/efficiency-gaming.png

1.50 frames per Watt at 1080p vs 4.1 frames per Watt for 5800X3D

im not the saying its inefficient here
if you have problem with that statement tell all the review sites like guru3d, Techpowerup or Anandtech

But what 12 games? Those are average with no max and min without absolute numbers.

[ChickenBalls]

as i said if you have problem with all that data you should email those review sites and complain about the inaccuracies. Maybe you will be able to convince them that the 13900K is as efficient as the 5800X3D

[xSOSxHawkens]

Publicado originalmente por ChickenBalls:

Flagships have needed 1kw PSU's for awhile now for real stability.

hmm I wonder who said that




https://www.tomshardware.com/news/1200w-power-requirement-rtx-4090

Recommended wattages should be taken with a grain of salt, since they won't apply to everyone's personal rig. Some might have a Core i9-12900K overclocked to 5.5GHz that consumes over 300W of power at its peak; others might be running a Ryzen 7 5800X3D that consumes less than 100W of power while gaming. Meanwhile, a Core i9-10980XE with a full blown RAID 10 setup and 10 hard drives could consume gobs of power. All of these combinations result in different levels of PSU power, regardless of what the AIB partner suggests.

I said that... Whats the point? Its accurate.

And we already covered the TPU lacks the ability to even test this stuff...

Again, why are you seemingly lacking the ability to understand all this? Go read up and educate yourself from some of the few publications that *can* test this stuff, I alreadly linked you a vid to one of them.

Publicado originalmente por ChickenBalls:

as i said if you have problem with all that data you should email those review sites and complain the inaccuracies. Maybe you will be able to convince them that the 13900K is as efficient as the 5800X3D

I'm not going to go lambast an otherwise good review site for having some somewhat incorrect views just because they lack the ability to accurately test the counter claims. Thats childish. The adult thing to do is to *understand* that there are *other* third party review publications that *do* have the equipment to test this stuff, and that for some specific cases (such as reviews of transient power spike impact on a user) going to one publication over the other is a wiser choice, as one knows what they are actually talking about from in house real world testing vs the other which is talking about theory or relaying told-to information.

Either that or go get your own oscilloscope and clamps and test it and prove it yourself. Otherwise have some respect for the professionals who have, and some grains of salt to the views of the publications who have not.

But one user to another, I can say I have seen the same things Jamesbonds and/or his friend have reported. I saw it with a quality 750 and my 4790k when I went to the Vega 64. Just like his bud, crashes under load etc. Was able to make it work by power limiting the card temp, but a 1kw solved it. My best bud, had a seasonic 850 with a 5600x/3060ti, went up to a 5800x3D and 7900xtx and thought he would be fine on his 850w. Was not, had crashes under load, fixed (temp) with power limit on GPU, fixed long term with 1kw PSU.

Its simple really.

[ChickenBalls]

fyi i subscribe to GN on Youtube and watch most of their reviews
and he has never said that 1000W psu is absolute minimum for a 4090 and powerdraw of the CPU doesn't matter which is what I have been trying to say. A decent 850W psu should be enough for 4090 and 5800X3D.

As the Tomshardware article clearly says that everyone has different rig and therefore different PSU requirement.

[Jamebonds1]

Publicado originalmente por ChickenBalls:

as i said if you have problem with all that data you should email those review sites and complain about the inaccuracies. Maybe you will be able to convince them that the 13900K is as efficient as the 5800X3D

CPU was not power factor calculated. So CPU can't be called out as efficient or inefficient unless someone have a way to put test equipment in between CPU and motherboard to find power factor. It can only be called massive power consuming.

Your source with "Energy Efficiency" does not claimed made to CPU standalone, but based on CPU only of benchmark software.

If you have watt meter, then you will see power is lower at idle CPU and GPU.