Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Which RTX is = 5500M
- Thread starter CM_1986
- Start date
- Sort by reaction score
AMD doesn't have a card that rivals the RTX 2080TI... and with regards to how Nvidia and AMD runs their GPUs... Ray tracing technology goes in one direction while metal or whatever AMD uses goes in another
It compares favorably against the GTX 1650. Maybe the RTX 2060 is kinda close? But yeah no ray tracing on the AMD card.
EDIT: just did some googling, nope, doesn't compete with the 2060.
The post you replied to were confused, and talked about desktop RTX 2080 Ti gaming cards, not mobile Quadra cards. Comparing with an RTX 2080 Ti is a mute point for many reasons, one being the thermal envelope, another being the ludicrousness of the 2080 Ti in the price/performance ratio aspect. AMD's cards are very competitive on price/perf and perf/power-consumption with the RDNA series.
The 5500M has 22 CUs I believe. The 5300M has 20 (RX 5300 has 22 as well, Radeon Pro has 20 - that's a noteworthy distinction).
The 5300M and 5500M actually aren't that far apart in most performance metrics though. Of course of you can use the 8GB of VRAM the 5500M with 8GB will significantly outperform the 5300M, but otherwise the difference isn't as huge as the naming suggests, and both outperform the Vega 20 in the last generation (in most tasks). Vega of course had very strong compute performance, so the difference is more striking in geometry based tasks like games, and less so in pure compute tasks, where the Vega is sometimes toe-to-toe with the 5300M.
For performance numbers, Max Tech and Max Yuriev (two channels same guy behind them) on YouTube has done tests on both cards. There aren't direct comparisons to the Quadros but if you have benchmark numbers for those you can compare yourself.
The RDNA GPUs are frankly excellent offerings
I was just trying to see how my Quadro RTX 5000 Max-Q compared to the 5500M 8GB.
My card is suppose to be on par with a desktop RTX 4000 for comparison.
The RTX cores are irrelevant at this point. RDNA and Vega can do fallback software ray tracing, which of course comes at a performance hit, but if ray tracing is necessary for a render job it's fine, and the perf∆ isn't actually that massive. Real-time ray tracing on RTX isn't even viable anyway, since it requires a significant reduction in ray casting to be real time, and really hurts performance, suggesting that the RTX cores aren't even fully hardware implementing ray tracing, and the shader pipeline still needs to do a lot of work calculating the ray. And with a well implemented Metal ray-tracer, (yes, Metal has ray tracing) AMD cards with their software-fallback actually perform quite well. There are also DXR implementations showcasing AMD ray tracing v Nvidia's. One made by Crytech and a Minecraft thingy.
But real time ray tracing isn't ready on anything yet. If you stick to 1080p, maybe 1440p, a desktop 2080Ti probably can be satisfactory, but basically nothing else.
RDNA2 will implement hardware accelerated ray tracing once it's more mature.
I couldn't find any info on the Quadra RTX 5000M.
However, the Quadra RTX 4000 desktop is listed as park performance of 7.1TFLOPs.
The Radeon Pro 5500M is listed for 4.6TFLOPs.
7.1TFLOPs seems very, very high for a mobile chip however. Are you sure the Quadro 5000M is equivalent to a desktop 4000?
you can't compare it's a different league
RTX 5000 Max-Q will be slightly slower than RTX 4000 both mobile and desktop version
I really wouldn't say that. I'd peg the perf∆ at, at most, 20% depending on task. But you're right that comparison can be hard, since they also have different feature sets. Like you've pointed out, the RTX has RTX and tensor cores, but the AMD card has a video block capable of h.265 10 bit HDR encodes, and better shader effeciency.
How is the RTX 5000 Max-Q beaten by a RTX 4000 Max-Q?
assuming it’s for CAD and FDE (Simulations).
I guess I’m trying to justify the MBP but I think I need to stick with the RTX.
there is 2 version in laptop: mobile and Max-Q (down clocked), mobile is more powerful
I can't understand why you want replace pc laptop with 2 times more powerful GPU than in MBP 16 with 5500M, there is no point unless you need MacOS
no, closer to 1650 or 1060 Max-Q
MBP 5500M 3DMark Time Spy graphics score = 3300, 3DMark FireStrike graphics score = 10300
1650 mobile 3DMark Time Spy graphics score = 3500, 3DMark FireStrike graphics score = 9300
1060 Max-Q 3DMark Time Spy graphics score = 3300, 3DMark FireStrike graphics score = 10400
1660Ti mobile 3Dmark Time Spy graphics score = 5600, 3DMark FireStrike graphics score = 14500
Last edited:
RX 5500M it's more powerful than variant 'Pro 5500M' available in MBP 16, so you can't be sure
RTX 5000 Max-Q is about 2 times raw power more, certified drivers (for CAD etc.), also h.265 10 bit HDR is supported
Last edited:
Whilst the RX 5300M and Radeon Pro 5300M diverge in the manner you describe, the RX5500M and Radeon Pro 5500M are listed by AMD as having the same performance in TFLOPs.
3DMark score doesn't confirm this
MBP 16 'Pro variant of 5500M' 3DMark Time Spy graphics score = 3300, 3DMark FireStrike graphics score = 10300
PC laptop with RX 5500M 3DMark Time Spy graphics score = 4000, 3DMark FireStrike graphics score = 1
11700
Do you see difference?
I don't know why Mac users can't do proper research before making valid point during discussion
From where do you have the numbers? I would a same a driver discrepancy. The Pro drivers are not targetted gaming and usually perform worse in gaming tasks on equivalent hardware, so if a pro driver was used that could be part of the explanation.
I did in fact do research, as I read the specifications for the two GPUs.
from online reviews, RX 5500M is rated TGP 85W, but Pro variant 5500M is TGP 50W, also difference in GPU clock and memory bandwidth