Next Generation Hardware Speculation with a Technical Spin [2018]

[borntosoul]

How long ago was the failed price points at $600.

Yes train people, it’s called marketing.
You and l get trained on a daily basis if you want to get deeper into it.
I global launch at $699 is not as crazy as it sounds, the early adopters pay and let the price drop when everything gets up to speed. As long as there is some content to justify the price people will pay, it’s got to be done properly, not some half assed effort.

Edit: a $400 consol 20 years ago is equivalent to $700 in today’s money.

 

[Xbat]

Yes but what does a extra $200 give you other than more faster storage and ram? I don't think it's going to get you much more TFlops. That's constrained by the available tech at the moment. So let's say you get 16 TFlops at $699 it gets you what? Games running at native 4k instead of temporal or Checker boarded 4k, well if the competition comes out at $450 16GB of RAM with 128GB fast nand to supplement the ram and 12 TFlops it will be the machine I get and many others I imagine, these consoles aren't released in a vacuum. Momentum is very important for new consoles/hardware that comes out especially in the age of social media we live in.

 

[borntosoul]

Agreed but everything is moving so fast.

That’s a great question... we should continue and ask ourselfs what does an extra $200 get you?
How much difference will it make, l believe it will be a lot, considering how tight the margins are.

That $200 extra could get an SSD drive? Or 32 gigs of mem? Or a faster GPU? The list goes on.........
Double the ram sounds nice though.

 

[Xbat]

What does double the ram bring though? 1TB nvme drive would be fantastic and would make the biggest difference. I feel they will be pushing the limits of an APU already to reach 12+ TFlops and it will come down to clock speed to get more out of the chips available and that becomes an issue in a small console type box.

 

[borntosoul]

Nvme drive on a consol with little overheads? Not sure how much difference it would make though. Someone could chip in and give a better explanation.

 

[borntosoul]

If you had to pick......
Faster and double the ram with a HDD.
Or...
Slower and half the ram with an SSD.

 

[dobwal]

That $200 extra could get an SSD drive? Or 32 gigs of mem? Or a faster GPU? The list goes on.........
Double the ram sounds nice though.

What’s the point of an SSD drive?

The moment you start a game, 90-95% (1 or 2 TB SSD) of the drive becomes useless. A fraction of flash could be used to provide the benefit that a SDD offers for a game in use. Also, expanding the amount of main ram or adding an intermediary level of slower ram would serve a game being played more readily.

 

[borntosoul]

Wouldn’t the sdd provide faster loading times for levels and worlds etc.

 

[Xbat]

I sure it would help tremendously with streaming and initial load times.

I agree though that a 128GB of flash and a big mechanical drive is what hopefully will happen.

 

[Shifty Geezer]

...the new machine should be at $599 and let the specs go to town. For all the tight asses that complain you can always stay on the last gen machines till the price finally goes down.

What timeline are you anticipating for the price drop? Dies aren't shrinking like they used to. Drives won't be getting cheaper than the cheapest available at launch. You might only be able to shed $150 over 4 years, meaning four years before it hits mainstream pricing and it may take a long time to get cheaper than that. Which means very marginal sales and not a healthy enough platform to attract the developers, so it'll get no software. Investments in first party games will cost more than they do now but will make far less money, so perhaps the entire venture would be lossy. You'd have to drag the generation out for longer.

And if your rival comes out with a $400 box, they'll sell gang-busters, get all the software, get the exclusives, and your product will be dead before it ever reaches mainstream pricing.

The only way a $600+ console would work IMO is as a higher-tier SKU, like a high-end PC graphics card in the same family. Already described, you could have a high-end SOC and a lower one that maximises profitability on yields. You keep the fairly vital $400 entry price to grow your platform so it isn't DOA, but also offer a better experience for the small hard-core niche and make the most money.

$600 only isn't going to happen - not because people need to be trained, but because it's a free-market with rival products at rival price-points.

 

[goonergaz]

How long ago was the failed price points at $600.

Yes train people, it’s called marketing.
You and l get trained on a daily basis if you want to get deeper into it.
I global launch at $699 is not as crazy as it sounds, the early adopters pay and let the price drop when everything gets up to speed. As long as there is some content to justify the price people will pay, it’s got to be done properly, not some half assed effort.

Edit: a $400 consol 20 years ago is equivalent to $700 in today’s money.

I think we only have to look at Xbox One X sales to see it won't work without significant reason to purchase.

Even going back to PS3 launch where people could buy it as a cheap Blu-ray player didn't help.

As suggested, a premium option might work but I personally think they would be better off repeating what happened this gen with a mid gen upgraded version.

 

[DieH@rd]

What would you be willing to sacrifice from the remaining pool of hardware specs to get 32 instead of 16?

Nothing.

I view 16GB of RAM as disappointingly low amount that would not enable creating of dramatic new nextgen experiences.

 

[itsmydamnation]

Nothing.

I view 16GB of RAM as disappointingly low amount that would not enable creating of dramatic new nextgen experiences.

What are these new capabilities that will be missed where you need more memory but not memory bandwidth?

 

[Gubbi]

I view 16GB of RAM as disappointingly low amount that would not enable creating of dramatic new nextgen experiences.

What are you basing that view on? Personal experience of massive image quality jump between existing 16GB and 32GB PC GPUs ?

Cheers

 

[goonergaz]

I think the problem is the perception that double last gen won't be enough to give a 'next gen' experience based on previous jumps.

I think even this gen folk suggested the power jump from PS360 wasn't enough either but clearly it has been.

We need to understand that Sony and MS are taking advice from Devs and therefore will not release a machine hamstrung by any one area.

 

[Shifty Geezer]

We need to understand that Sony and MS are taking advice from Devs and therefore will not release a machine hamstrung by any one area.

I don't see the logic in that statement, I'm afraid. The limiting factor isn't engineering skill or awareness of requirements, solved by including developers. It's money, and the cost of components. Devs can give advice on what they'd like to see, but if the price point and timelines limit what can be released, all the developer input in the world can't make up for a 'weak' generation. If to get a generational advance in visuals requires 32 GB and 24 TF (I don't believe it does, same as you), and lal the devs tell MS+Sony this, the hardware still won't achieve that because of cost.

 

[borntosoul]

What timeline are you anticipating for the price drop? Dies aren't shrinking like they used to. Drives won't be getting cheaper than the cheapest available at launch. You might only be able to shed $150 over 4 years, meaning four years before it hits mainstream pricing and it may take a long time to get cheaper than that. Which means very marginal sales and not a healthy enough platform to attract the developers, so it'll get no software. Investments in first party games will cost more than they do now but will make far less money, so perhaps the entire venture would be lossy. You'd have to drag the generation out for longer.

And if your rival comes out with a $400 box, they'll sell gang-busters, get all the software, get the exclusives, and your product will be dead before it ever reaches mainstream pricing.

The only way a $600+ console would work IMO is as a higher-tier SKU, like a high-end PC graphics card in the same family. Already described, you could have a high-end SOC and a lower one that maximises profitability on yields. You keep the fairly vital $400 entry price to grow your platform so it isn't DOA, but also offer a better experience for the small hard-core niche and make the most money.

$600 only isn't going to happen - not because people need to be trained, but because it's a free-market with rival products at rival price-points.

What $400 consol released soon will make a difference to the Xbox one that’s out now?
Either gotta wait till 2021 or charge more and release earlier, that’s the way I see it.
Mememory pricing should go down and storage too, and hopefully a die shrink.

 

[Shifty Geezer]

What $400 consol released soon will make a difference to the Xbox one that’s out now?

A $400-$450 console released 2019-2020 will be a very significant advance on XB1. This whole thread is about how to get maximum gains within the price bracket!

Either gotta wait till 2021 or charge more and release earlier, that’s the way I see it.

If the plan is for it to be $400 by 2021 and launch soon (so end of 2019?), you're needing a $200 price drop in two years. This gen has seen a $100 drop in three years. PS3 saw a massive price drop only because the hardware was lossy and Sony bled money, and last gen stayed pricey, from $400 to $200/ $500 to $250 over 7 years. One shouldn't expect $200 in two years when the rate of cost reduction is constantly decelerating, unless there's a new tech coming that'll solve that.

Mememory pricing should go down and storage too, and hopefully a die shrink.

Memory pricing will go down. Storage won't below its minimal price unless you use a high-end storage solution. Putting $100 of SSD in there could see a $50 reduction in a couple of years for the same capacity. Silicon is very unlikely to reduce, so no thermal reductions or system shrinks. Certainly not enough to save $200 in total.

And if the only thing you can price reduce is storage (SSD), you can easily support two SKUs at launch, or not bother and have it user upgradeable.

That's all ignoring the competition aspect with your rival's mainstream priced box outselling yours 4:1. Same reason the GTX 1060 outsells the 1080 10:1.

 

[borntosoul]

Hard to compare pc hardware because a consols specs are set, so more likely the higher will be programmed for.

lve been around long enough to know the fastest system doesn’t always win the race. It takes everything, timing , software, marketing, etc to be successful.
Still think $400 is redundant pricing closing in on 2020.

 

[goonergaz]

I don't see the logic in that statement, I'm afraid. The limiting factor isn't engineering skill or awareness of requirements, solved by including developers. It's money, and the cost of components. Devs can give advice on what they'd like to see, but if the price point and timelines limit what can be released, all the developer input in the world can't make up for a 'weak' generation. If to get a generational advance in visuals requires 32 GB and 24 TF (I don't believe it does, same as you), and lal the devs tell MS+Sony this, the hardware still won't achieve that because of cost.

I probably didn't say it the right way, I'm bad enough as it is and these days restricted to using my phone which is a PITB!

I just meant they would use Devs advice to spend the budget as 'wisely' as possible. Whilst I appreciate in the example given Devs may ask for 32gb which might not be possible (for financial reasons) but if Devs insisted 16gb would be bad next gen then Sony/MS can at least make an informed choice and decide to cut a less needed feature, take a loss or risk a high release price.

I probably didn't help clear up what I'm trying to say lol