They've simply said that with a 7GB/s SSD (already available) and a 2:1 compression ratio (the same as claimed by both Sony and Microsoft) you will see a 14GB/s output by doing the decompression on the GPU.
Neither Sony nor Microsoft are claiming decompression on the GPU.
Are people really so doubtful that multi-terraflop GPU's can do the same job in compute shaders as the consoles likely quite cheap hardware blocks?
Well for starters, what use are floating point operations for file decompression?
And if so, where's the evidence supporting these doubts?
I wrote this several times, but let's try it again.
How Zlib works.
Zlib (or ZIP, the most popular compression algorithm), like the vast majority of compression formats uses
single threaded decompression.
It basically counts known sequences of bits and groups them together by giving them a different "name". As a
very simple example of compression:
"0000000 111111 0000 1111" can be compressed into sequences of "number of zeros"-"number of ones", of which you could say it's 7x0 + 6x1 + 4x0 + 4x1, or "'111 110 100 100".
With this, I "compressed" 21 digits into 12.
But the result is one sequential file of which you can't change the order or take random blocks out of, otherwise it becomes unreadable. Or as explained
in the blog post:
Old fashioned compressors like Zip parsed the compressed bit stream serially, acting on each bit in different ways, which requires lots of branches in the decoder - does this bit tell you it's a match or a literal, how many bits of offset should I fetch, etc. This is also creates an inherent data dependency, where decoding each token depends on the last, because you have to know where the previous token ends to find the next one. This means the CPU has to wait for each step of the decoder before it begins the next step.
To make ZIP compression levels higher, you need bigger files. To make ZIP decompression parallel, you'd need to split the original file into smaller files. So to make ZIP decompression more parallel, you lose compression ratio, and effective IO throughput in the process.
So at least with Zlib or anything else that uses ZIP, you don't gain by adding more threads to it.
GPUs aren't better than CPUs at running single-threaded code. They excel at highly parallel tasks. Which is why making a GPGPU decompressor for ZIP makes no sense, other than to save CPU cycles if the raw I/O is slow enough so the GPU decompression doesn't become a bottleneck. The only way to make Zlib decompression faster than a CPU is to make a dedicated hardware block for it.
Kraken is a very different compression format that was developed for
decompressing on 2 threads. It's not great for parallel work, but it's better than Zlib's one thread. But it's still not going to gain ridiculous amounts of performance from a GPU.
Which is why, to get crazy high Kraken performance like an >8GB/s output, once again a dedicated hardware block is needed. Unless the game engine is consistently trying to load tens or hundreds of textures at the same time (which I don't think it happens). But decompressing one texture is always going to be faster on a 3GHz CPU than a 1.9GHz shader processor.
Cerny wasn't going to spend a third of the PS5 hardware presentation talking about the importance of their high-performance decompressor if the problem could have been solved with a couple more CUs on the GPU.
This is why nVidia, by claiming "
we'll just use our many parallel TFLOPs on this mostly single-threaded problem that uses INT operations" is sounding shady as hell.
And the fact that they're not even disclosing what compression format is making their GPUs so damn effective at decompression makes it even shadier.
And it's not below nVidia to be shady (or lie or be dishonest) on features that are promised years in advance. Nor is it below AMD or Intel BTW. This isn't vendor-specific.
They're not lying if their GPGPU decompressor only outputs 14GB/s if it's loading 10000 zlib-compressed textures in parallel, even if it never happens in a real-life scenario. They're just being dishonest.
Their marketing team knows how much popularity and attention the fast decompression features on Microsoft and Sony's consoles have garnered, and it would be harder to sell their $700-$1500 graphics cards if they had nothing to say about it.
