Unreal 5 Demo vs Series X Reveal

[Pemalite]

SvennoJ said:

Thanks, I learned something new about SSDs :)

My confusion however mostly came from the HDD write speed being significantly faster than its read speed. Is that just the luck of the draw, was it writing at the outer orbit while reading from the inner orbit. HDD afaik have constant angular velocity, thus the outside moves much faster than the inside.
Or does it have to do with caching, HDD says thank you for the data and does more work without holding up the test tool.



Second, why are random reads so much slower on SSD. Seek time should be instant, yet sequential reads are still 3.6 to 50 times faster. It actually writes the smallest chunks faster than it can read them. That looks like managing your data chunks is still as important as before. (And probably why, for example, it still takes KSP a long time to load, tons of little files)

Ah looking up what the Q and T stand for helps a bit, Queue depth (number of requests at once) and Threads used. I guess Q1T1 is stress testing the IO controller mainly, all separate requests which the IO controller can add together to write more efficiently, yet while reading I guess the tool waits for the data before sending the next request.

Anyway not as simple as dump a ton of files in a directory and hope for the best. Optimizing for efficient reading of data will still help.

The reason for the write speed being larger than the read speed is that essentially the hard drive is doing two things.
* It's reading directly from disk and not the hard drives Ram.
* It's write speed is mis-reported, it's reporting the speed it writes into the hard drives Ram and not to the disk itself.

The reason why the random reads are slower on an SSD is actually really simple...
It's much easier for an SSD's controller to read and access "pages" of data that is next to each other rather than access it from completely opposite ends of the SSD, the data request doesn't have to travel as far.

The further the data is away, the longer distance that data has to travel hence impacting seek times, it's the exact same issue a mechanical drive has, just on a much smaller scale.

And the more requests you have the more contention you have in the SSD's controller, it can only do a finite amount of tasks at a time due to processor and memory limitations.

eva01beserk said:

I dont mean RT is bad, just that on the rtx series is bad, or I would say experimental. Nvidia says they will get 4x the rt performance on Ampere. Im pretty sure they will still be better than amd, but even amd should beat the rtx line. Maybe it was a necessity as devs had something to work with and now nvidia and even amd could work from that.

Or maybe I should ask, will the 2080ti hold up to the 3000 series as well as the 1080ti did to the 2000's? Acording to mores law is dead if people are looking to upgrade and RT is a must for you that you should ignore the 2080ti and a 3060 should be better.

But like you said, cant do much with leaks. probaably wont know untill rdna 2 releases, but nvidia already confirmed that RT wise, they will smoke the 2000's

It's still the best form of consumer hardware Ray Tracing we have, AMD doesn't have Ray Tracing, Ampere, Playstation 5, Xbox Series don't even exist on the market.

But you are correct, it's only first generation stuff, more improvements will obviously be made, the PC never stops improving.