News on Nvidia’s site, the 5000 series is coming out at the end of January, according to this official link the performance of the RTX 5090 in D5 render is more than doubled. If someone gets it before and can test it it would be good if you could confirm this performance for us.
Yes, I also have the 4090 line, but the performance they advertise justifies the change to the 5090.
Before, going up a step in graphics cards only improved speed by 20 to 30%, but now they advertise more than double the speed in D5, if this is really the case, it is worth making the change.
Seeing the reports from some tech advisors, the new Nvidia RTX 5090 doubles its performance with applications that can use DLSS 4.0, increasing the number of FPS by generating fake frames.
This improvement is useful for video games and their FPS performance, but in D5 this improvement would only be seen in the visualization of the scene in real time.
My question is whether the performance that Nvidia talks about only refers to this FPS performance and not to the render time, which is lower and closer to the usual 30% in each Upgrade.
We will have to wait for the first ones who have the 5000 line card and test the render times with D5 Benchmark.
Bad News with the new Nvidia 5000, at least so far, the Passmark of the Nvidia 5090 is only 5.85% higher than the RTX 4090, on the PassMark Software site.
Since I started using my first card, the RTX 980, I always consult this site, which is the reference that recommends the Lumion site, and they have always been reliable in their Benchmarks.
Because of this difference in performance, I don’t think it’s worth upgrading to the RTX 5090 except for live scene navigation.
I’ll wait for some user to publish the D5 Benchmark and confirm the performance.
There are differences of opinion regarding software benchmarks. Some software benchmarks show real gains in rendering, such as Unreal Engine. Other tests are not as exciting. But I think we have to look at each case individually to see if it is worth it. In the case of D5RENDER, we will only know for sure when someone does a review, which I don’t think will happen any time soon. I have a 4090 and I’m thinking about upgrading, but only in about 3 months, when the price and availability stabilize.
You’re right, it was so low that I hadn’t seen it, I always thought that the RTX 5090 result would be above, surpassing the 4090.
It’s incredible that it’s below the first result of the RTX 4090 by 18 places.
If this is so, the only improvement in the TRX 5000 line is in the FPS with the frames added by fake frame interpolation, with the only improvement in the real-time visualization of the scene.
Going from the RTX 3090 to the RTX 5090 seems like a justifiable jump despite what the RTX 5000 line seems to perform like.
When the use of the RTX 5090 is extended to several users, then we will know the true performance in D5 render.
Also, the results in the benchmark are very variable, you can see people with the RTX 4090 in first place with 19:09 seconds in the 2K render and also with the same card more than 30 pages back with 1:43 minutes of render time.
It seems that each performance depends on many variables and not just the card.