Next-gen PS5 and next Xbox speculation launch thread |OT5| - It's in RDNA
- Thread starter Mecha Meister
- Start date
- OT
- Status
The high end is ridiculous expensive, but there are now all in one solutions that are modest in their pricing. ANd as Hoo-doo pointed out it will get cheaper as time goes on.
Right now it's still trying to figure out what people are willing to spend. i think Sony right now has the best more streamlined take on VR. What ever the new controllers/vr headset is, it's not going to be as expensive as Valve's and it will be priced hopefully close to original PSVR if not cheaper. The good thing is PSVR is super cheap now with tons of bundles and will work with PS5.
There are more VR titles coming, and you will see in the next year or so more publishers go in on VR in a big way.
I really hope PS5 continues with Sony's regular tradition of stability updates. This forum won't be anywhere near as fun without them.
That's not what anyone was complaining about. People are asking for a next gene first party game that was built from the ground up for next gen, not just an XBOne game that was pretted up. No amount of enhancement can make a current gen, a next gen title. Adding more pixels isn't going to make a PS3 game play like a PS4 title.
I'm willing to bet on 60CUs at 1800MHz for 13.8tflops.Okay,, on request by i-Lo here is the repost of the most hated table in the last thread ...
With some modifications to the range it covers ...
Nope, Xbox is VEGA
This thread would be much more enjoyable if we just chill and stop regurgitating at least half that list.
First priority: Stop the fake, member-created pastebins "for lulz"
The Endless Cycle Committee appreciates the new thread even while the committee is on vacation!
KTHXBAI
KTHXBAI
We don't need to see any Navi card using HBM2 to validate its inclusion in any of the next-gen APUs. All it requires is a change in the controller. Kinda like how at the start f this both Sony and MS had very different memory solutions albeit both using the same Jaguar/GCN arch.
The missing black border below Clock in MHz is killing my OCD.
We will see in some days I'm pretty sure your like most of many others in that thread that playing the "concern game" are very highly wrong ;) but was expected how would end that thread . And btw is very very likely exactly what happened last time and what will happen to this year Sony exclusive
We've had launch timing, announcing timing, and power polls in the previous threads. I think a poll on price or total RAM would be good, if we're going to do a poll.
Second person on the sony side and third person on the whole next gen front that has alluded t the longevity of the current gen way into the life of next-gen consoles.
Is there something these people are saying that we should take more seriously?
Last edited:
Mostly that is due to the new arch.
- Previously performance was measured based on CU count. (64sp/CU) and those CUs in turn connected to Shader Engines and a maximum of 4 SE/GPU for a max of 64CU.
- But now there are rumors that there could be more SEs/GPU and its at least ben confirmed that we now have 1.5x perf/watt. We can no longer use the GCN CUs to directly measure performance
I just take it as PS5 will be BC (which we know) and that teams are confident PS4 software sales will be strong on PS5. I expect them to switch from making Pro patches to PS5 patches.
Maybe Sony does packaging similar to Xbox 360 games shipped after Xbox One BC update and has PS4 games like say Yakuza 7 ship in a different color box and a label that says PS4 and PS5
Like blue boxes for pre-PS5 games (like today), grayish black boxes for PS4 games with PS5 support and dark black (like UHD movies) for PS5 only games
To be fair I would ban all of that crap talk from this thread and i would also add Xbox using next next gen tech as well. We all know it only takes obe rumour or a bit of unfounded speculation to make most folk lose their shit.
The more things change, the more they stay the same.
Don't remember the Sony comments, but from Sony's side it's as simple as "please don't stop buying PS4s right now" rather than something truly indicative of what kind of first party support PS4 will be getting a year or two into the ps5's life.
Japanese devs (especially with yakuza type games which sell decently but aren't mega sellers) for the most part I expect to be in a cross-gen phase that is similar to the last gen transition. Unless of course, PS5 does better than the ps4 in sales in Japan.
We will see the AAA publishers that have games which appeal to a wider audience moving onto next-gen only, a lot faster though this time around. Where AC Unity was a rare 3rd party AAA next-gen only title 1 year after this gen started, I expect to see many more 3rd party AAA games being next-gen only in around that timeframe.
A lot of technical stuff that won't really mean much to anyone per se, like support for 64bit, the introduction of shader engines to better handle scheduling to the CUs, the expansion of the number of shader engines present, support for things like tesselation and lossless color compression.......etc. Basically covering a span of GCN1 through to GCN 5.
The only major thing we know so far about RDNA is that there are now 8 shader engines (or at least this is rumored)...... that's double the number of shader engines on GCN5 and that will mean that the GPU front end in the RDNA arch has been changed which could also mean the schedule has been improved along with the size of the L2 and L3 caches. It could also mean (if they can use the same amount of CU/SE as in GCN) that they can have up to 128CUs this tie around (i doubt this though).
I think the reason RDNA is more efficient is mostly due to better task scheduling which could b a direct by-product of increasing the number of SEs.
The discussions in this thread are much more interesting when you start with the assumption that both teams behind the next-gen consoles are comprised of competent engineers and business people who are making good and valid choices about what these products will be based on their overall platform strategy and that that strategy makes objective sense. Anyone not capable of doing this is very likely to be too invested in the success or failure of one or the other party to contribute anything useful to the discussion. Once you start from that point of view, it becomes a lot easier to evaluate the validity of these rumors and whether they should be given any weight. It's a lot better method than the, "this rumor suggests an outcome that I want, so I'm going to choose to believe it" method.
Can we also stop when one poster says something treating it as some movement? "I" is not "we". I know in the console war mentality you have to choose a side and everyone not on your side is the enemy and should be grouped as "other", but maybe recognize how gross that is and how that type of thinking manifests in other areas of conflict in society. Don't be a console-ist.
Can we also stop when one poster says something treating it as some movement? "I" is not "we". I know in the console war mentality you have to choose a side and everyone not on your side is the enemy and should be grouped as "other", but maybe recognize how gross that is and how that type of thinking manifests in other areas of conflict in society. Don't be a console-ist.
E3: No detailed specs for consoles by MS but detailed specs for PC NAVI GPUs by AMD.
Appreciate the explanation. Curious about the second point.
I understand that more than 4 SE/GPU renders old basis for measuring the power obsolete. However, how does improved efficiency regarding power consumption act to make CU obsolete unless the 1.5x improvement is greater than what would have been possible just from 16nm to 7nm shrinkage?
If both are launching within a month of each other ~Nov 2020 then no, it'll be next year for detailed specs. If one comes earlier than the other (and the other knows this) then maybe more than say Scorpio level reveal..?
We will get full Navi details for the PC in E3.
I still expect a PS and Xbox Meeting in early 2020.
That's exactly it..... the 1.5x improvement is greater than what the shrink would have given. Especially when looking at the Vega 7 GPU. But there is also the 1.25x IPC in addition to that too. So not only is the GPU performing better with much lower TDP its also doing more than what a similarly specced GCN GPU will have done (1.25x more to be exact). You know all that AMD TF vs Nvidia TF talk? Well, now that gap is smaller.
So now we can't just take CU and multiply stuff by stuff and get a TF number based on what we have been doing with GCN all this time, we have to factor in something else.
To me what is really interesting is the increase in SEs from 4 to 8. That is actually a big deal if true and could explain a number of things. Also could make a number of things possible. Keep in mind, GCN5 had a maximum of 16 CU/SE. This tech could go either way.
Expecting:
PS5 GPU weaker than 2018 $500 RTX 2070 card
PS4 b/c limited to 1st party and loot box
because game license doesn't carry over to new platform unless it continues to make money
Plus continues to decrease in value. It will drop to 1 game every 3 months in 2022 and online storage increase to 20GB but no USB save backup
PS5 GPU weaker than 2018 $500 RTX 2070 card
PS4 b/c limited to 1st party and loot box
because game license doesn't carry over to new platform unless it continues to make money
Plus continues to decrease in value. It will drop to 1 game every 3 months in 2022 and online storage increase to 20GB but no USB save backup
Very well said! These devices are being built by some of the smartest engineers and designers in the industry and they're both going to be amazing.
I'm very interested to see how each company prioritized the various aspects of the components in their design. I think on thing that makes consoles so interesting is the constraints placed on the design, they have to do so many things with so many limitations - I'm always impressed by what these companies come up with.
Me neither..... think the ceiling is 48CU clocked at 1800Mhz. For 11TF that feels like GCN5 13.5TF.
I'm a PSVR owner. Regret is the best word I can use to describe it. Same goes for the PRO but that's another subject. HOWEVER better screen, wireless, and better tracking and there is no comparison to the type of experience you can have with VR vs regular gaming. Price may be an issue, but it's not the biggest issue. Even if PSVR was $150-$200, I couldn't see the value proposition. Sales would increase but plateau pretty quickly. It's niche, it's "new", it's needs tech that isn't available yet.
Believe what you want to believe.
I find the reddit dev summit leak more credible than anything posted in these threads.
- AMD said Navi clocks higher
- Navi is 7nm
- PS4 Pro has 36 CUs on 16nm
I feel like both gens being x86 makes cross gen a thing much easier for studios to do than during previous transitions. If the titles themselves aren't particularly ambitious (in a hardware dependency sense) then the could probably cut away at game and still have a serviceable version for PS4/XB1 that still looks/plays pretty well.
I don't think developers have ever had the guarantee of "base level" hardware being as high as it will be next gen, so there might be a sizeable period where they can stretch their legs and get a feel for what they are able to do. Things they've been wanting to that they never could, and things none of us have even come up with yet. I get a feeling that GDCs a few years from now are gonna be extra awesome. But until they figure those things out, a lot of studios will still design games and levels etc the way they always have, just at higher fidelity.
I am curious about people's stated positions on who will be more powerful and, most importantly, why. (I will list my bias' below so people know where I am coming from on a personal level should they wish to temper my words with my perceived personal tastes.) I will state that I think it will take a major gaff of some kind to produce any major shift. Sony had the PS3 and MS had the Xbox One. Mistakes I expect are too recent to have been forgotten and repeated in some form.
From my perspective Sony is in the drivers seat. For MS to launch at roughly the same price/ power point would be suicide. As MS, you would be banking on an even hardware contest which moves the battle for market/mind share to exclusives/ price/ and services lineup. Isn't that just asking to get stomped? Even if you like the studio acquisitions by MS, their next games are still unknowns as to their success. If I was MS, I wouldn't count on Ninja Theory producing my new "HALO" and saving my generation.
Sony does not need to disrupt the market and take large risks. (Note: This does not mean they would not, or should not, take an early loss on the PS5.) Pick you price point and subsidy. $400-500 with a subsidy of probably $60-100. Realistically speaking, there are points of major diminishing returns. Finding those "bang for the buck" points is probably a better determining factor than a pure $/ mm^2 approach.
Now if you are MS and you have to disrupt the normal course of events, what do you do? Sony can probably produce a nice value proposition and count on their exclusives to carry them. But if you are MS, how do you ensure a notable power advantage? This really cannot be something minor like: 5% more TF and an extra 500GB of SSD space. That isn't going to move the needle much. At least, I cannot see how it would. You would have a good idea of what is possible on 7nm, early Navi and Zen 2. So how do you leapfrog your opposition without pouring money into it, selling it at a horrible loss, and receiving a 15% or 20% TF upgrade at most?
Basically: How would you attack the situation if you are MS? The only answers I have been able to come up with so far revolve around leveraging a later launch date (but within a year certainly, and 6 months preferably.) If your opponent is going for late 2019 (which well believed rumors early pegged as the initial release date, with a delay for varied and vague reasons), Zen 2, early Navi, 7nm and an SSD (with whatever customization's Sony makes to accommodate their design goals - i.e. - backwards compatibility, RT, etc) - then you target early Zen 3, mature Navi, 7nm+ (I think this was expected to be 10% on clock speed and 20% on density gains over the non-EUV 7nm - feel free to correct me if this is wrong),
Could you launch Lockhart first and have Anaconda 6 months to a year later? Or would that be conceding the power high ground too much to your opponent?
My biases - I want competition. Hardware differentiation would be nice if possible. The only existing platform specific franchises which have any hold on me are Horizon: Zero Dawn and some vague hope that HALO returns to form. Neither of those are likely to heavily influence which console I get. To date, my console bias that is relevant here (ignoring old Nintendo, Atari, Sega, etc) is I owned an original Xbox, a 360 and then the PS4 (my first ever Sony console). I don't regret any of those decisions.
From my perspective Sony is in the drivers seat. For MS to launch at roughly the same price/ power point would be suicide. As MS, you would be banking on an even hardware contest which moves the battle for market/mind share to exclusives/ price/ and services lineup. Isn't that just asking to get stomped? Even if you like the studio acquisitions by MS, their next games are still unknowns as to their success. If I was MS, I wouldn't count on Ninja Theory producing my new "HALO" and saving my generation.
Sony does not need to disrupt the market and take large risks. (Note: This does not mean they would not, or should not, take an early loss on the PS5.) Pick you price point and subsidy. $400-500 with a subsidy of probably $60-100. Realistically speaking, there are points of major diminishing returns. Finding those "bang for the buck" points is probably a better determining factor than a pure $/ mm^2 approach.
Now if you are MS and you have to disrupt the normal course of events, what do you do? Sony can probably produce a nice value proposition and count on their exclusives to carry them. But if you are MS, how do you ensure a notable power advantage? This really cannot be something minor like: 5% more TF and an extra 500GB of SSD space. That isn't going to move the needle much. At least, I cannot see how it would. You would have a good idea of what is possible on 7nm, early Navi and Zen 2. So how do you leapfrog your opposition without pouring money into it, selling it at a horrible loss, and receiving a 15% or 20% TF upgrade at most?
Basically: How would you attack the situation if you are MS? The only answers I have been able to come up with so far revolve around leveraging a later launch date (but within a year certainly, and 6 months preferably.) If your opponent is going for late 2019 (which well believed rumors early pegged as the initial release date, with a delay for varied and vague reasons), Zen 2, early Navi, 7nm and an SSD (with whatever customization's Sony makes to accommodate their design goals - i.e. - backwards compatibility, RT, etc) - then you target early Zen 3, mature Navi, 7nm+ (I think this was expected to be 10% on clock speed and 20% on density gains over the non-EUV 7nm - feel free to correct me if this is wrong),
Could you launch Lockhart first and have Anaconda 6 months to a year later? Or would that be conceding the power high ground too much to your opponent?
My biases - I want competition. Hardware differentiation would be nice if possible. The only existing platform specific franchises which have any hold on me are Horizon: Zero Dawn and some vague hope that HALO returns to form. Neither of those are likely to heavily influence which console I get. To date, my console bias that is relevant here (ignoring old Nintendo, Atari, Sega, etc) is I owned an original Xbox, a 360 and then the PS4 (my first ever Sony console). I don't regret any of those decisions.
Yup, I have to agree with wit you.
I think the issue is that most people tend to allow their own personal bias get in the way of things. Makes people less reasonable lol.
Personally, for me, I don't think that that power crown is anywhere near as important as a lot of people here seem to think it is. And I think MS knows that too. At the end of the day, it comes down to games and price. And that's the bottom line. Console warriors will talk up power day and night and pixel and frame count, but I simply don't believe the power difference that will exist between these two consoles to be enough to make it really matter.
And I think MS is showing its hand already. They have been snapping up studios and making a 1080p/1440p version of their next-gen console. Those two things alone are what my theory is based on..... the games and Price.
People think MS wants to make a "premium console" and a cheaper one. I don't think so, simply because that is not important at all in the grand scheme of things. I think MS wants to make a 4K console that is the same price as the PS5 but still like 2% - 10% more powerful. Even if that means they take a slightly bigger loss on the hardware. And then have a console that comes in at $100 less. That is the best two SKU strategy that I can imagine them doing, the one that will make the most sense and have the best all-around effect. And if some random guy like me on a forum can think of it, then I am sure they can too.
If they actually succeed in doing that is another matter.
Last edited:
My uncle works for pastebin:
It mentions HT-ICE, so I googled and I don't know what this tool does:
VR link seems to be the new connection designed vor vr.
Would be massive if ps5 is 12,7 TF and mx 11,3 tf. I am still team 10tf for both.
It mentions HT-ICE, so I googled and I don't know what this tool does:
VR link seems to be the new connection designed vor vr.
Would be massive if ps5 is 12,7 TF and mx 11,3 tf. I am still team 10tf for both.
- Status