Xbox Series X [XBSX] [Release November 10 2020]

I got an email from the Microsoft Store that said "Owner's Series" & it congratulated me on being one of the first to join the next-generation of Xbox.

Nothing else was in there about when it would ship. Just gave me a link to get to know my console & then links to pre-order accessories & games.

Tommy McClain
 
I received Walmart Canada’s confirmation email upon placing the order, and the estimated delivery day says Nov 17th, I hope that’s not accurate :(
 
Don't get twisted though, Xbox is here as well! We've got that extra thicc #SeriesX

alongside his smaller brother, the #XboxSeriesS

! Got the whole gang back together again
2764.svg

 
Isn't dual stream just implying vertex and pixel shaders? That the multicore command processors are just more efficient at handling them than the the prior GCN architecture. I haven't kept up with the 'tech' 'tech' stuff happening at AMD, so my memory on their architectures history is quite hazy.

@Jay actually answered this but though I'd quote so you saw it:

Dual command processor - run different level of DX, one for game and a lower priority one for OS.

(That's my understanding too. I think the X1 had something similar in principle for games and dash overlay / snap).

Edit: funnily enough, just come across this in the stream I'm going through (link is timestamped):

 
Last edited:
@Jay actually answered this but though I'd quote so you saw it:



(That's my understanding too. I think the X1 had something similar in principle for games and dash overlay / snap).

Edit: funnily enough, just come across this in the stream I'm going through (link is timestamped):


The wording threw me a bit. Didn't equate dual stream to dual command processor.
 
Last edited:
So does L2 then go to infinity cache and the infinity cache to the memory controllers? Or are we looking at something closer to esram setup, where it's being deployed as a scratch pad that is not developer accessible?
I'm not sure about the exact relationship between the L2 and Infinity Cache. It appears like one possible route for data is L2 to Infinity Cache, but there are options for controlling whether one or the other is used at the driver level at least.
The ESRAM was managed by developers, since they needed to control when targets or buffers were brought in or out of it. Resources like the framebuffer could even be split so that parts were left out, if size was a problem.

Isn't dual stream just implying vertex and pixel shaders? That the multicore command processors are just more efficient at handling them than the the prior GCN architecture. I haven't kept up with the 'tech' 'tech' stuff happening at AMD, so my memory on their architectures history is quite hazy.
Stream would probably be the command buffers for the overall GPU. In GCN, it was usually something the block could only handle from one source at a time. The current gen had two command processor blocks to handle two the system and game in parallel with less concern of latency or throughput problems because one client or the other was blocking a single processor. RDNA introduced a command processor that could support two graphics queues, which may be related to what Microsoft was discussing.
 
Stream would probably be the command buffers for the overall GPU. In GCN, it was usually something the block could only handle from one source at a time. The current gen had two command processor blocks to handle two the system and game in parallel with less concern of latency or throughput problems because one client or the other was blocking a single processor. RDNA introduced a command processor that could support two graphics queues, which may be related to what Microsoft was discussing.
I think they mentioned "virtualized command streams"? So it's probably one [multi-core] command processor with context switching.
After all, dedicating a whole piece of hardware just for the dash seems like a waste.
 
I think they mentioned "virtualized command streams"? So it's probably one [multi-core] command processor with context switching.
After all, dedicating a whole piece of hardware just for the dash seems like a waste.
Pretty sure it's 2 distinct command processors, with custom firmware with different DX levels.
 
I think they mentioned "virtualized command streams"? So it's probably one [multi-core] command processor with context switching.
After all, dedicating a whole piece of hardware just for the dash seems like a waste.
The PS4 and Xbox One had two command processors, possibly because the unit could not be fully virtualized or because context switching at the time didn't meet their quality of service limits.
RDNA's dual queues could be managed within the same virtual instance for one thing, but it's also the case that since the current gen consoles were launched that GCN and RDNA introduced complete virtualization support with multiple VMs.

The command processor block has contained more than one processor for many GPUs, pre-GCN even. RDNA's block may have added another to better handle context switching and multiple-queue functionality.
 
I am a bit perplexed that we barely see anything of the Series X's features promoted or demonstrated in actual exclusive games.

For the Playstation 5 we see games running and interviews talking features implemented in actual games.

The Series X is promoted as a machine that has these technologies and plays games, but barely games demoed

Because they were waiting for the full RDNA 2 featureset. I.e making sure things were being implemented as needed on AMD's GPU line and the silicon being ready, the APIs needing to be ready along with that, etc. Also MS restructuring the SDK into Gamecore has had a very big impact on timeliness of devkit updates (and thus what features are there for devs to leverage).

We are seeing some of these features leveraged in some of the 3P games though, and updates like the Gears 5 optimized. Does it really matter if the game is exclusive or not showing off these features right out of the gate? It's a nice bonus, but the 1P content will be there. Sony had the advantage of finalizing their spec earlier and having a lot more cross-gen support on the 1P than first assumed, meaning since their PS5 devkit (from what others are saying) is like a "supercharged PS4 devkit environment", it's probably been a lot easier for devs to get off the ground running with that. Combined with, again, a lot of their 1P at launch and launch period being cross-gen (Spiderman, Horizon, most likely Demon's Souls etc.), I'm not surprised their 1P have had time to leverage some of these features at launch.

Also it's a bit disingenuous to say barely any games have been demoed. 1P? Yeah, to some extent, though again they've shown off Gears 5 with the new updates. There's been a decent amount of 3P games shown on the platform now though: Scorn, The Medium, AC Valhalla, Watch Dogs Legion, DiRT 5, etc. There've been demonstrations, just not enough 1P next-gen content for some folk's tastes (including my own). Hopefully there's a treat or two in that regard in November, we'll have to wait and see.
 
Also MS restructuring the SDK into Gamecore has had a very big impact on timeliness of devkit updates (and thus what features are there for devs to leverage).

It is bewildering that for the second generation launch in a row, Microsoft are seemingly behind Sony again on the firmware/SDK readiness. The world's biggest software company who literally make VisualStudio, which is what devs use for PlayStation development.

Just... how? Was the demise of XDK and the rise GDK a very recent decision?
 
It is bewildering that for the second generation launch in a row, Microsoft are seemingly behind Sony again on the firmware/SDK readiness. The world's biggest software company who literally make VisualStudio, which is what devs use for PlayStation development.

Just... how? Was the demise of XDK and the rise GDK a very recent decision?

Sony just has to worry about PS5. MS is trying to reduce the work needed to go from xbox to pc as much as possible while bringing a new api in dx ultimate .

Aside from that they are starting to show off the third party games now with ray tracing and othe features
 
It is bewildering that for the second generation launch in a row, Microsoft are seemingly behind Sony again on the firmware/SDK readiness. The world's biggest software company who literally make VisualStudio, which is what devs use for PlayStation development.

Just... how? Was the demise of XDK and the rise GDK a very recent decision?

That's a good question. I can only guess that the talks about the Xbox division being pigeonholed as a whole regards R&D during the 2015-2017 period is more true than first assumed. That may've affected their ability to pace next-gen console development as quickly as they've of preferred, including whatever plans they had for redesigning the SDK for better integration with PC.

Would be neat if we got a write-up at some point on the state of Xbox during those years before Phil Spencer got the promotion and convinced Satya Nadella to invest big into the brand again. When you look at the timing of the initial acquisitions (Ninja Theory, Obsidian, etc.), it's a bit wild to think that was only two years ago.
 
Late 2019 to Early 2020 going by release notes of GameCore.

Perfect timing, and a great decision by the people who 1) wouldn't have to do it and 2) wouldn't have to use it during 3) the run up to launch of a complex product up against the market leader.

I'm sure it helped that there wasn't fully functioning silicon over this period too. :yep2:

Edit:

"But sir, I don't have a parachute on yet!"
"Have to get your first jump out of the way some time, laddy!"
 
Back
Top