NVIDIA GeForce GTX 1660 Linux Benchmarks

[Dedale]

Interesting. Those tests came fast.

[birdie]

Interesting results. I'm just glad about AMD's fine wine drivers and the 580 being able to compete with a new Nvidia card.

I wonder how much performance would change and what the performance per dollar would be if Michael could undervolt the GPUs and test them.

With my RX 580, simply lowering the last 3 states by 25mv was enough to make it run around 15C lower under a full load which stopped thermal throttling, lowered the fan noise level, and improved its gaming performance.

With almost maxed out settings, at 1080p Metro 2033 Redux went from 60 fps smooth for one minute of game play when throttling would kick in at 75C causing dips 40-45 fps to playing it with 60 fps with dips down to 50-55 fps in the more GPU intense areas while only hitting 65C without needing the fans to kick in full blast.

Absolute most people out there do not undervolt their GPUs, so this mantra that AMD "shines" when being undervolted is just a stupid excuse to justify very poor AMD's showing for the past three years.

Also, undervolting is a huge lottery - works for some and doesn't work for others.

[aaahaaap]

Keep in mind though that particularly if using Radeon hardware with newer Vulkan games and using the RADV driver, that's where the NVIDIA Linux driver performance tended to have a much better outcome.

What do you mean by this? I'm not seeing this in the graphs. Don't you mean it the other way around?

[bridgman]

The frames per second per watt graphs would have been interesting for Hitman and Paraview

[theriddick]

I can only assume that nvidia is no longer making series 10 dies anymore and has fully moved onto their 12nm process for all things, which is why we are seeing such cards as this as a replacement for older gen stuff. There isn't a huge leap in performance but it means nvidia can forget about 16/14nm?

[Mike Frett]

Watch those ASUS cards. I've had issues with their heatsinks not being properly seated and the fans locking up after six months.

[skeevy420]

Absolute most people out there do not undervolt their GPUs, so this mantra that AMD "shines" when being undervolted is just a stupid excuse to justify very poor AMD's showing for the past three years.

Also, undervolting is a huge lottery - works for some and doesn't work for others.

While I'm aware it's anecdotal, undervolting my previous two AMD GPUs helped with performance more than anything. Under gaming loads, my 260x would hit 90C w/o undervolting (75-80C afterwards) and my 580 went from 75-85C to 65C. Both instances prevented throttling...yeah, that 260x ran really hot...surprised it lasted 7 or 8 years...

I'm also just as curious of the effect that undervolting would have on Nvidia GPUs. Since throttling would make my 580 have a 45/60 in the min/max FPS tests versus the 50/60 I have afterwards, makes me wonder if Nvidia cards are in a similar situation after a bit of use.

For fairness sake, I imagined some sort of "gaming" motherboard with an automagic undervolting option being used (do those exist? I know automagic overclock exists) -- anyhoo, whatever that comes up with is what the tests would use.

Also, coin mining made undervolting rise in popularity. It's safe to assume more people are undervolting and bios modding now than they ever were.

[yurikoles]

Michael any change to have performance per dollar comparison of Radeon VII with rest modern GPUs?

[Rulet]

Why is not said which linux distribution was used and hardware in this graphic card test? How testing can be serious if distribution(and software which was used) and hardware is not mentioned? Or maybe I missed something?

[Michael]

Why is not said which linux distribution was used and hardware in this graphic card test? How testing can be serious if distribution(and software which was used) and hardware is not mentioned? Or maybe I missed something?

You missed it, all the hardware/software information is clearly displayed on a table on the second page.