NeoGAF’s member ‘winjer‘ has compiled the latest version of the BulkLoadDemo, which now supports DirectStorage 1.2 with GPU Decompression. This DirectStorage 1.2 Benchmark will give you an idea of what your storage devices are capable of, so we suggest giving it a go.
As said, the latest version of Microsoft’s Bulk Load Demo now supports DirectStorage 1.2. By default, the demo uses GPU Decompression. However, according to its GitHub page, you can enable CPU decompression by using the “-gpu-decompression” argument. In order to enable CPU decompression, you must use the custom “-gpu-decompression {0}” command line.
You can download the compiled version of this benchmark from here.
DirectStorage promises to reduce loading times on all storage devices. While many assume that this technique only works with SSDs, it will also work with hard-disk drives. Basically, everyone will benefit from it.
As we’ve already reported, Ratchet & Clank: Rift Apart will be the first game to support DirectStorage 1.2 with GPU Decompression. As Nixxes stated, traditionally, this decompression is handled by the CPU, but at a certain point there is an advantage to letting the GPU handle this, as this enables a higher bandwidth for streaming assets from storage to the graphics card. Thus, Nixxes uses this to quickly load high-quality textures and environments with a high level of detail.
It will be interesting to see how Ratchet & Clank: Rift Apart will perform on PC. For instance, while Marvel’s Spider-Man Remastered did not have any stutters on PC, it had really high CPU requirements. These high CPU requirements were mainly due to the lack of support for DirectStorage. Despite that, Nixxes was still able to introduce additional CPU optimizations via a number of post-launch updates.
Ratchet & Clank: Rift Apart is coming to PC on July 26th. That’s ten days from today. So, until then, you can download this DirectStorage 1.2 Benchmark and give it a go.
Stay tuned for more!
John is the founder and Editor in Chief at DSOGaming. He is a PC gaming fan and highly supports the modding and indie communities. Before creating DSOGaming, John worked on numerous gaming websites. While he is a die-hard PC gamer, his gaming roots can be found on consoles. John loved – and still does – the 16-bit consoles, and considers SNES to be one of the best consoles. Still, the PC platform won him over consoles. That was mainly due to 3DFX and its iconic dedicated 3D accelerator graphics card, Voodoo 2. John has also written a higher degree thesis on the “The Evolution of PC graphics cards.”
Contact: Email
idk what this does, but results were better for me on hdd than sata ssd
DF said they’re interested in testing Ratchet and Clank with the base level hardware requirements which are extremely low as well as with newer hardware to see how it all shakes out and to determine just how much needs to be cut to enable that instant world switching on such low level hardware
My RTX 3070 Ti decompresses at a rate of 2.01 GB/s, and my Ryzen 7 3800X decompresses at almost exactly the same rate. That being said, with CPU decompression the rate seems to vary slightly, whereas with GPU decompression the speed remains consistent. Since games tend to use a lot of CPU time anyway, I definitely see GPU decompression as advantageous just to ensure consistent load times of CPU usage and the type of CPU and number of cores.
That variation on the CPU side likely stems from the so-called “balanced” power-profile of Windows.
Setting that to “performance” should reduce it.
And keep in mind that GPU decompression costs resources, which could drop the FPS during gameplay.
This was my active power plan at the time I ran the benchmark (and at all other times as I have Process Lasso locking this power plan as active so it can’t change):
https://uploads.disquscdn.com/images/114c3782805543eb75adf92831ba22c92580a0d08817783dc010d99421a8cecd.png
And yes, GPU decompression does use GPU power, but for only 2 seconds in that test. Granted in a real video game it would be longer due to the amount of assets a game uses being significantly more than a small test like that. That being said (and I forgot to check earlier) I’m assuming they’re not using the 3D cores in the GPU, but rather probably one of the compute cores (tensor cores, optical flow accelerators, CUDA cores, etc) in order to minimize the effect on FPS.
would be nice direct storage in returnal, do you think it can be implemented in a old released game like this john?
I don’t expect developers to revisit older games in order to implement DirectStorage. In theory, with a lot of work, it’s possible. In reality, most developers won’t do it.
8.63 GB loaded in 0.64s, 13.46 GB/s.
Pretty impressive.
2TB Sabrent Rocket 4 Plus Gaming (direct storage optimised firmware):
23.45 GB/s // Loaded 0.37s
8TB Sabrent Rocket 4 Plus:
19.97 GB/s // Loaded in 0.44s
8TB Samsung 870 QVO SATA connected to a USB-C 3.1 Gen 2 enclosure):
2.10 GB/s // Loaded in 4.11s
Not too shabby.
My 990 Pro loaded in 0.40s.