The only logical conclusion is Sony are using both a controller from Samsung and Phison. One for the SSD, one for ReRAM 😉
So many folks getting stuck on the minutiae when in reality these two systems will come out, be leaps and bounds better than last console transition, everyone will get to play amazing games and the PC master race will continue to turn it's nose up to the boxes.
Percent here, a flop there, we're all gonna be having fun.
Man the comments there, it seems people just do not understand what a fast loading drive can fully achieve. Its not just how fast a game loads, but also how everything can be put into place and how much you can have on screen at one time. All systems having SSDs will benefit all next-gen though.
I strongly doubt that sony (or anyone for that matter) would go with a custom proprietary likely more expensive option for just a 0.5-1sec advantage. It makes zero sense.
From where did those 2 infos get revealed and leaked for PS5? I missed both.
So many folks getting stuck on the minutiae when in reality these two systems will come out, be leaps and bounds better than last console transition, everyone will get to play amazing games and the PC master race will continue to turn it's nose up to the boxes.
Percent here, a flop there, we're all gonna be having fun.
Isn't Elite Series 2 already using USB C as well? Shape of the port really plays no role in what the port really can offer.Because ever since the allegedly DualShock patent, people took it for granted. Even though it can be very well using a USB A to C cable.
Yes. Both the controller and dock use USB C.Isn't Elite Series 2 already using USB C as well? Shape of the port really plays no role in what the port really can offer.
That's clearly a DualShock 3, not 4. Looks like they're canceling the PS5 and just reissuing the PS3 with an SSD.The thing that's funny is that they're assuming Sony is using Samsung because of a DS4 in the product video and in the Phison video for their controller, they actually use a PS4 Pro and DS4.
No, the GitHub leak is concrete info. It can be interpreted in multiple ways, but any description of the consoles has to include a reasonable interpretation of it. It can't just be dismissed the way many pastebins can.The only concrete info we have is what MS and sony have revealed themselves.
There's another, more logical way to reconcile everything. We know the GitHub notes were from tests of backwards compatibility. The current example we have for Sony doing that is on PS4 Pro, which natively has 36 CUs at 911 MHz. Its PS4 compatibility mode downgrades that to 18 CUs at 800 MHz to perfectly mirror its predecessor machine. But it also has Boost Mode, meant to smooth out dynamic res and unstable framerates of games that don't officially support the Pro hardware. That mode runs 18 CUs at 911 MHz. That is, keeping the CU number the same as the prior hardware, but bumping up the clocks to current max. (I suspect it also disables the ID buffer and other new Pro hardware as superfluous, as suggested by Sony patents.) Note that this setup doesn't represent the total capacity of the GPU in the console, just the max they allow for boosting.Yes I know. I just meant that my guess atm until I learn otherwise is that it could well be set apart from the other GPU resources or something along those lines. That would explain why we hear about 'TF numbers don't tell whole story' as well as 9.2 TF from the AMD data stress tests and still could dovetail with what insiders have said more or less. I'm trying to find ways to reconcile the various bits and pieces of the puzzle so far. :)
Exclusives will show the most difference, sure. But I don't know where the idea that third parties never have big gaps comes from. (It's a common belief, I'm not singling you out.) Did third parties balk to make Call of Duty much lower resolution and lower framerate on Xbox One versus PS4 right at launch? Did they neglect to make Red Dead Redemption or Far Cry 5 or many others much better on One X than Pro? History shows us that third parties take advantage of the hardware as much as they can. Multiplatform titles are competing with other games on the individual platforms, at their respective power levels. Not just the lowest common denominator.I don't think we will see that big of a gap with 3rd party titles. Now 1st party, say for instance, Spiderman 2. Yeah that will definitely show the difference.
iPad Pro also uses C-C and it's annoying as hell.You are right. Pixel might be an exception then ? Never saw any other smartphone charger with USB C - C before.
Only because of the ubiquity of USB A. C needs to become more popular. Surprised it's taking so long. USB C is so much better.
Is it safe to assume OsirisBlack is not getting verified? Hecht
Isn't Elite Series 2 already using USB C as well? Shape of the port really plays no role in what the port really can offer.
I prefer A to C that way pretty much everything can be compatibleOnly because of the ubiquity of USB A. C needs to become more popular. Surprised it's taking so long. USB C is so much better.
It's basically just memory paging, except the storage is much faster so it emulates RAM much better than slow ass HDDs."They'll be able to use it as virtual memory" how is 2-4 GB/s anywhere close to the 13 GB/s that RAM can do.
Isn't Elite Series 2 already using USB C as well? Shape of the port really plays no role in what the port really can offer.
"They'll be able to use it as virtual memory" how is 2-4 GB/s anywhere close to the 13 GB/s that RAM can do.
Cheaper to make does not mean they end up with the cheaper solution. It could just as easily mean "cheaper than such a solution ould have otherwise cost". But said the solution could likely still end up being more expensive.Again, part of Sony's patent is about making the SSD cheaper to make, so in this case proprietary isn't going to mean it's more expensive than using a third party.
The issue with that leak is that it is without context. That's like a company saying we built the fastest vehicle we ever made and no one knowing that all they make are tractors.No, the GitHub leak is concrete info. It can be interpreted in multiple ways, but any description of the consoles has to include a reasonable interpretation of it. It can't just be dismissed the way many pastebins can.
You are wrong, probably because you are basing your theory on dated information.USB A & C are just connector types. The XSX controller takes a USB C connector as well, but the front-facing port on the console is USB A.
The connector type does not dictate the speeds or other properties of the port. Presumably the USB A ports on the XSX are USB 3.1 spec, which is the spec people associate with the USB C connector.
See this article for more background/clarification: https://www.newnex.com/technology-articles-USB3.1-vs-USBTypeC.php
What if part of the focus on SSD is developing super-fast/optimized paging algorithms on the software/OS-side? Any patents filed on that?The faster your storage, the less RAM you need to use for "buffering", aka storing data that may or may not be needed but has to be held in RAM "just in case".
Would you be surprised? Anyway, it's a great time to be alive. To see some truly awesome consoles.Hopefully there isn't anyone on this forum that says "PC Master Race" seriously.
The faster your storage, the less RAM you need to use for "buffering", aka storing data that may or may not be needed but has to be held in RAM "just in case".
No I get but the super fine details of something is 0.000X faster or whatnot is silly to argue or get bummed out about because the leap from this gen to next gen is going to be significantly more substantial then the leap from previous gen to now. That is cool.
Then they can partition over more of the SSD storage to the system. Right? Just like an extra 16gb or soIt's still 6x slower than RAM and would require people having parts of their SSD unused (which is going to be a tough ask as games next-gen will be 100+ GBs, have to be installed, and their SSDs will probably be 1 TB).
Agree, and I think Liabe did an excellent job explaining that at in that post.We can massage the information to suit whatever agenda we are pushing at the time, but it doesn't change that said information is without context. We don't even know if its complete or how old it is.
what did he said exactly?
A huge difference can be seen in the suspend/resume of games. Transferring all that RAM to the SSD and back can create a big gap.Even SATA SSD speeds are more than fine, we're talking milliseconds difference now lol.
vs the current 250x+ slower than RAM HDDs we are currently using?It's still 6x slower than RAM and would require people having parts of their SSD unused (which is going to be a tough ask as games next-gen will be 100+ GBs, have to be installed, and their SSDs will probably be 1 TB).
What if part of the focus on SSD is developing super-fast/optimized paging algorithms on the software/OS-side? Any patents filed on that?
How will they benefit.
I doubt developers are going to put a ton of assets in a spot that you can zoom past at a million miles an hour.
Sonic Team discovered pretty quickly that this does not go well.
I'd encourage you to read gofreak 's take.I only understand roughly half of these words, soooo... anexanhume, any thoughts?
At the moment, Sony won't cop to exact details about the SSD—who makes it, whether it utilizes the new PCIe 4.0 standard—but Cerny claims that it has a raw bandwidth higher than any SSD available for PCs. That's not all. "The raw read speed is important," Cerny says, "but so are the details of the I/O [input-output] mechanisms and the software stack that we put on top of them. I got a PlayStation 4 Pro and then I put in a SSD that cost as much as the PlayStation 4 Pro—it might be one-third faster." As opposed to 19 times faster for the next-gen.
It's orders of magnitudes slower than GPU cache, which is what matters for performance. No persistent storage solution is fast enough for what they are talking about. This is very much some "power of the cloud" nonsense.
Let's put things in perspective, these SSD solutions being talked about are crowing about an optimal read speed of 50 GB/s, which is a pretty unrealistically optimistic prediction. GPU cache access speeds can reach 2 TB/s, that speed is orders of magnitude faster. Even when not accessing GPU cache, you're still talking about hundreds of GB/s from VRAM. Persistent storage is much, much too slow for these kinds of processes. Things like VAOs for LOD systems need to reside in the absolute fastest bits of memory, as they are frequently accessed.
I'm talking about the uses of virtual memory they theorize in the video are nonsense. Of course I'm not saying virtual memory is nonsense, I've written my own paging system from scratch in the past.
Virtual memory is not good for parts of graphics systems, your graphics technologies need the absolute fastest memory access of any part of your computer.
A huge difference can be seen in the suspend/resume of games. Transferring all that RAM to the SSD and back can create a big gap.
You're suspending the gamestate that is in RAM to the SSD so you can play another game. Then come back exactly where you were before. A savestate for modern games.Using SSD storage as ram is one of the dumbest things I've seen recently.
No that won't work, be happy load times will be a second or 2 at most.
This should be a good bookmark post for when Digital Foundry comparisons start rolling in.It goes beyond that, I/O is important for data streaming. If the SeX is only 2-3GB/s and the PS5 is 7-10GB/s then depending on how third parties use the SSDs we may see SeX games have lower quality assets to make up for the slower I/O.
Its frankly pretty bad if the I/O gap is that big.
Oculus Quest is USB-C to C