Hacker News Reader

Comment by leakycap | Hacker News Reader

A NVIDIA 2080 graphics card from 2018 still surpasses the M5 for gaming. The M5 Pro coming early next year will likely finally catch up with the 8-year-old 2080.

I'm happy to hear your games work well for you, but it sounds like the games you're playing aren't demanding compared to modern AAA titles. Most games released in the last year or two don't run well on my 2080 test system at anything approaching decent graphics.

View on HN

A 2080 is about the same performance as a 5060 and every game is going to be able to run on a 5060. You might not be running it at 4K Ultra with ray tracing enabled but you should be able to run at like 1080p High or better.

Whether or not the M5 GPU is actually capable of that level of performance or whether the drivers will let it reach its potential is of course a completely different story. GPU performance is hard to estimate based on raw specs, you just have to run benchmarks and see what you end up with.

> A 2080 is about the same performance as a 5060

A 5060 outperforms a 2080 by roughly 20% on most titles, across the board, not cherry-picking for the best results. They are not about the same.

> you should be able to run at like 1080p High or better

This is disconnected from reality. 1080p low/medium, some games are playable but not enjoyable. Remember, I actually have a 2080, so I'm not just guessing.

> GPU performance is hard to estimate based on raw specs, you just have to run benchmarks and see what you end up with.

Rich coming from someone who claims a 7 year old graphics card is "about the same" as a card which has 2.5x better RayTracing, has 3x faster DLSS, faster VRAM, and much better AI capabilities. The 2080 can't even encode/decode AV1...

> 5060 outperforms a 2080 by roughly 20%

Is this a typo? I’m surprised the difference is so small after 3 generations.

It's much higher in some categories, but in general across gameplay the 2080 is only about 20% slower than the 5060 in otherwise similar systems. NVIDIA's 3000 series was mostly worse than the 2080 except the 3090, which itself is still is incredible compared to today's 5xxx cards

> This is disconnected from reality. 1080p low/medium, some games are playable but not enjoyable.

Is it? Most people care the game is fun which is unrelated to the settings you use (framerates do impact enjoyability though for fast paced games).

Even visually, the difference between settings hasn't been significant for decades at this point.

But does it weight 7lbs and sounds like an airplane on the tarmac while you play and looks like it was designed by a 13yo that listens to Linkin Park?

You can absolutely play demanding and modern AAA titles on MBPs.

What you can't do is run them expecting to have every detail knob maxed out and very high framerates.

Yet blind test after blind test shows that most people can't even fully appreciate extra vs medium details.

A reminder that the majority of what people actually play isn't "modern AAA titles": https://steamcharts.com/top

> I'm happy to hear your games work well for you, but it sounds like the games you're playing aren't demanding compared to modern AAA titles.

That's why I made the specific distinction in the comment you're responding to

When a $599 Windows laptop with a 3060 can play AAA titles and your $1599 MBP can't, I wouldn't normally call that great for gaming.

> A NVIDIA 2080 graphics card from 2018 still surpasses the M5 for gaming.

How much energy does it burn while surpassing the M5?

Yeah, I totally hear you. People are always like "please, make sure my desktop computer only uses 30 watts of energy" and "I'm an eco-gamer, please turn off the RGB lights and let's all play the game on eco/low graphics settings, kumbaya"

> please turn off the RGB lights

This but unironically ;-)

Reply on news.ycombinator.com