New Steam hardware inbound

[kksnowbear]

31 minutes ago, LucShep said:

What me and others were talking was specifically about how Valve designs and supports their devices compared to generic prebuilts - not about anyone’s shop, business, or personal philosophy.

Uhhh, @Dragon1-1 was talking about my shop, and customers. You were pretty much the only one talking about how Valve wonderfully designs and supports their devices.  And maybe they do, not an area of extensive experience for me.

My comments are primarily geared toward the hardware.  And that absolutely *is* an area of extensive experience for me.  The best software will never run well (if at all) on a crappy hardware design.  These Steam Machines are small.  The smaller a device gets, the more likely it's going to have problems with heat, unless performance is cut back.  Need we discuss Intel's recent missteps?

BTW Please be reminded, "hardware" is literally in the title of the thread, and the hardware I'm discussing is absolutely, 100% on topic.  FWIW it's not the Steam software or "ecosystem" I'm concerned with; not their support or "transparency".  It's the fact that cute little machines always suffer limitations in proportion to their size - a concept with which I am very familiar.  And there are some rules that Valve, for all their transparency and reputation, simply cannot change.  Small and heat are enemies.

Edited 19 hours ago by kksnowbear

[LucShep]

42 minutes ago, kksnowbear said:

Uhhh, @Dragon1-1 was talking about my shop, and customers. You were pretty much the only one talking about how Valve wonderfully designs and supports their devices.  And maybe they do, not an area of extensive experience for me.

My comments are primarily geared toward the hardware.  And that absolutely *is* an area of extensive experience for me.  The best software will never run well (if at all) on a crappy hardware design.  These Steam Machines are small.  The smaller a device gets, the more likely it's going to have problems with heat, unless performance is cut back.  Need we discuss Intel's recent missteps?

No, you’re also evaluating this as if it were just a tiny Windows PC - but that’s not what Valve is building.

You're experienced with system building, sure. But how many systems have you actually built and tuned around SteamOS?
Because the performance and thermal expectations simply aren’t the same.

SteamOS ≠ Windows.
It doesn’t carry the same legacy overhead, driver baggage, or thermal assumptions.
Valve controls the entire software stack end-to-end - it’s a purpose-built, highly refined Linux environment, prepared by themselves for their device.
And that fundamentally changes the hardware requirements.

It’s the same reason the Steam Deck delivers both stable performance and good thermals, on a fraction of the power that a Windows handheld would need.
People have tried Windows 11 on it, tuned and debloated to the bone, and it still performs noticeably worse.

So when you bring up heat concerns, you’re applying the logic of generic prebuilts or small Windows boxes. That’s simply not the design model here.

And this loops back to the original point: we were discussing Valve’s track record in integrated hardware + software.
Not your shop or your clients.

The Deck already proved that a tightly validated Linux system can run efficiently, reliably, and without the usual small-form-factor headaches.

That’s why this Steam Machine concept could become genuinely disruptive, a possible new branch of PC gaming - even if the specs are very far from top tier.
It doesn’t need high-end hardware or blistering performance to succeed. It just needs to deliver the right experience for its target audience - mainstream and casual gamers.
 

Edited 19 hours ago by LucShep

[kksnowbear]

13 minutes ago, LucShep said:

So when you bring up heat concerns, you’re applying the logic of generic prebuilts or small Windows boxes.

Uhhh, again, you're assuming a ton.  My experience includes embedded systems, IPCs and SBCs, that have nothing to do with Microsoft, let alone windows.  iRMX, MultiBus, etc, so on...and yes, Windows (and before that, DOS).

Guess you didn't know that, though.

But that's getting off topic.  The relevant point is that, in any event, the rules of heat/size apply...always have, always will.  Pesky physics.

Edited 18 hours ago by kksnowbear

[LucShep]

12 minutes ago, kksnowbear said:

Uhhh, again, you're assuming a ton.  My experience includes embedded systems, IPCs and SBCs, that have nothing to do with Microsoft, let alone windows.  iRMX, MultiBus, etc, so on...and yes, Windows (and before that, DOS).

Guess you didn't know that, though.

In any event, the rules of heat/size apply...always have, always will.  Pesky physics.

Nobody’s disputing physics - small boxes have thermal limits, sure. We’ve already covered that.

My point is simply that Valve isn’t building a tiny Windows PC.
Just like an Xbox Series X, a PlayStation 5/Pro, or a Nintendo Switch 2 aren’t “tiny Windows PCs” either.

SteamOS lets them run games efficiently on much lower power budgets than a general-purpose Windows system, which is exactly why the Valve's Steam Deck performs as well as it does in such a small chassis.

Physics still applies, of course - but the performance targets and the software stack are completely different, and that matters when judging the design.

So the real question is: what specific design problem are you assuming exists here that somehow escaped Valve’s own engineers during research & development?
 

Edited 18 hours ago by LucShep

[MAXsenna]

58 minutes ago, kksnowbear said:

Tell me: From your view, what's "wrong" with trying to put informed perspective, based on relevant experiences, where concerned parties will actually see it? 

Nothing. But this back and forth bickering won't make anyone, except us that are still here, read through it.

 

59 minutes ago, kksnowbear said:

After all, we have to consider the many readers who never post anything, and may not even be forum members.  Are the lurkers somehow less important?

No, but if one is in the camp that actually consider this to play DCS, they would have stopped after your first post, if they ever bothered to read it. Even when it was quite informative, it was quite lengthy. And the answer was already given, it will be bad for DCS, in one of the previous posts.

Like @Dragon1-1 writes. DCS players are not the target. Pretty sure most will drop by Linus and Jay first when they test it. Maybe even Steve, while I doubt it. They won't have the attention span Steve provides.  

[Dragon1-1]

59 minutes ago, kksnowbear said:

Uhhh, @Dragon1-1 was talking about my shop, and customers.

No, I was talking about how the overwhelming majority of Valve's customer base doesn't come to your shop. You seem to act as if Valve should consider your customers in particular as the center of the universe. It's OK for you, but it sure isn't what Valve is doing, or should be doing. Steam services more customers than you'll ever get to see in your lifetime, spread across more places than you (or any other human) will ever get to visit in your lifetime. That calls for a somewhat different approach than a one person brick and mortar store, don't you think?

13 minutes ago, kksnowbear said:

In any event, the rules of heat/size apply...always have, always will.  Pesky physics.

That doesn't mean there's no room for improvement versus Windows SFF devices, though. Since you worked with so many different architectures, you probably know that Windows is not exactly optimized for low heat operation, low overhead or smooth operation on low performance hardware. Linux often is the choice for older hardware for a reason, it doesn't suffer from bloat and assumptions that latest Windows releases make.

Will this enable cramming a 5090 into a shoebox sized SFF? Probably not. Will this enable a usable gaming machine to be built inside one? It might. Again, consoles exist, they don't routinely melt their plastic casings, and most people who buy them don't end in your shop, or any other similar one. You're thinking of it as a PC first, but again, it's a console that can play most (but not all) PC games.

Edited 18 hours ago by Dragon1-1

[kksnowbear]

2 minutes ago, LucShep said:

what specific design problem are you assuming exists here that somehow escaped Valve’s own engineers during research  & development?

It doesn't have to escape them entirely to cause issues.  They can go too far in the direction of small, and/or use components that are power efficient but lack performance.  Happens all the time, even among *highly* qualified engineers.

One very big problem is engineers have bosses, and the bosses are worried more about cost.  So the engineers often wind up doing stuff they know better than, to save their jobs.  Again, happens all the time.

[kksnowbear]

10 minutes ago, Dragon1-1 said:

most people who buy them don't end in your shop, or any other similar one

You're right...a lot of them just wind up bitching when it shuts down and misbehaves all the time because of heat, then eventually bug mommy and daddy to the point they'll spend (another) $500 so they don't have to hear it anymore.

And that's just one group.  It's a separate group that finally realizes that little machines will always have proportionally more problems with heat.  They're the ones that are now happily running PCs built by skilled builders, rather than going back to the source of the problem again.

I'm also wondering if any of you have seen the type of 'maintenance" these gamers do on their machines.  Hint: They don't.  Doesn't help the heat equation.  Small computers are also more likely to suffer from lack or/poor maintenance.

Edited 18 hours ago by kksnowbear

[scommander2]

I am wondering that if there is a poll for electing Steam hardware or PCs for DCS..., mmm..., the result will be very interesting... 

[kksnowbear]

27 minutes ago, MAXsenna said:

Nothing. But this back and forth bickering won't make anyone, except us that are still here, read through it.

I hate to seem disagreeable, but I can't imagine how you'd have any way of knowing who reads what exactly.

I guess you'd be surprised to learn that I have been approached by complete strangers, more than once, and wound up gaining a new client, strictly based on my presence in gaming/simmimg forums.

[LucShep]

1 hour ago, kksnowbear said:

It doesn't have to escape them entirely to cause issues.  They can go too far in the direction of small, and/or use components that are power efficient but lack performance.  Happens all the time, even among *highly* qualified engineers.

One very big problem is engineers have bosses, and the bosses are worried more about cost.  So the engineers often wind up doing stuff they know better than, to save their jobs.  Again, happens all the time.

Again, nobody’s denying that bad engineering decisions can happen anywhere.
But there’s a major point you keep brushing aside and contradicts yours: Valve has no shareholders to appease. 
That single fact changes everything about how this should be evaluated.

Instead, you should look at this "Steam Machine" from a different analytical perspective (i.e, see the "bigger picture") instead of just focusing on the tiny little box "hardware perspective". I'm pretty sure then you'll understand why all the commotion around it, why this is such a big thing. Not "how it ventilates" or "how many FPS for the dollar"...
Some points to digest that are very, very relevant:

Steam Store = instant ecosystem + zero effort.
To get games Valve doesn’t need to negotiate ports, pay for exclusives, or convince publishers to support their platform.
Every game that runs on Steam already “supports” their hardware!
No other console maker has that luxury/advantage.

Money + stability.
Valve is private, absolutely and absurdly profitable, and under zero quarterly pressure. No forced release cycles, no investor panic, no market-share chasing.
They move at their own pace and only release hardware when they feel it's right.

Steam Deck experience.
The Steam Deck gave them real hardware R&D, a mature SteamOS/Proton stack, controller tech, and millions of users stress-testing compatibility for them.
Something nobody else has.

Telemetry data nobody else can match.
Valve sees the entire PC landscape in real time: GPUs, CPUs, framerates, screen sizes, what games people play, what “good enough” performance looks like, how often people play handheld vs docked.
Sony/Nintendo/MS-Xbox only see their own walled gardens, all limited to their own platforms, and nothing as broad as the entire PC gaming landscape.

Zero existential risk.
Valve could release a Steam Console, Deck 2, mini-box, GPU dock, VR box - whatever. And none of these decisions risk collapsing their business.
You see, even if a Steam PC Console only sells moderately well, every sale boosts SteamOS, Proton and Linux support from devs, every sale pushes people deeper into the Steam ecosystem. They benefit even at medium scale. Basically, they win even without “winning”.

Sony/Nintendo/MS-Xbox can’t afford a flop. Valve can, and still profit from it indirectly. 
 

Edited 17 hours ago by LucShep

[kksnowbear]

34 minutes ago, MAXsenna said:

Even when it was quite informative, it was quite lengthy

Well, sometimes the best answers aren't the easiest, and sometimes there's just no substitute for determined legwork and putting in the time.

Those in some way bothered by the length might simply consider not reading it.

Edited 18 hours ago by kksnowbear

[Dragon1-1]

25 minutes ago, kksnowbear said:

And that's just one group.

Yeah, it's "just one group". Sure. So, how many millions of PCs have you built and sold this year? 

BTW, PS5 isn't particularly notorious for running hot. Of course, sometimes they break, and they sure aren't receiving maintenance, quite the contrary, they stand in one place and build up a bunch of dust. Doesn't change the fact they sold 75 million units. I'm pretty sure you didn't. 

What I'm trying to tell you is that your second group is simply ignored by Valve in their design and marketing of Steam Machine, because people who rely on skilled builders number in tens and not in millions. You're basically a local cobbler trying to judge Nike by your own personal yardstick. There's a reason people buy handmade leather shoes over Nikes, and there's also a reason Nike doesn't deal in handmade leather shoes. What is not in question, however, is the relative size of both customer groups.

Edited 18 hours ago by Dragon1-1

[kksnowbear]

13 minutes ago, Dragon1-1 said:

Yeah, it's "just one group". Sure. So, how many millions of PCs have you built and sold this year? 

BTW, PS5 isn't particularly notorious for running hot. Of course, sometimes they break, and they sure aren't receiving maintenance, quite the contrary, they stand in one place and build up a bunch of dust. Doesn't change the fact they sold 75 million units. I'm pretty sure you didn't. 

What I'm trying to tell you is that your second group is simply ignored by Valve in their design and marketing of Steam Machine, because people who rely on skilled builders number in tens and not in millions. You're basically a local cobbler trying to judge Nike by your own personal yardstick. 

I never questioned Valve is bigger than anything I do.  You seem fixated on that.

The thing is, being big doesn't at all guarantee they're right.

Edited 18 hours ago by kksnowbear

[MAXsenna]

22 minutes ago, kksnowbear said:

I hate to seem disagreeable, but I can't imagine how you'd have any way of knowing who reads what exactly

You don't. But have very different views on this obviously. Just by the numbers of users here that ask the same questions over and over again. All the wishes that have been wished before. All the double reporting of bugs. I could go on. It shows me that users don't even bother to search the forums, or even Google it. (Now they are in the minority, here). But it fits very well with my user experience doing IT enterprise work, and for some reason I always end up as "the head of support". Most users are like this, and that's fine. And I'll give most users credit that they won't even be thinking about getting this console for DCS anyway. So, I'd wager that most users that would consider it, or at least check if it is feasible. They won't get past those 10 first posts.

22 minutes ago, kksnowbear said:

guess you'd be surprised to learn that I have been approached by complete strangers, more than once, and wound up gaining a new client, strictly based on my presence in gaming/simmimg forums.

I'd be more surprised if it didn't happen actually. Of course a user that is knowledgeable enough to one's limit knowledge, and read your posts, and that you actually "advertise" yourself in that regard. Why wouldn't they approach you?

I even said so myself!  

Edited 17 hours ago by MAXsenna

[Dragon1-1]

18 minutes ago, kksnowbear said:

I never questioned Valve is bigger than anything I do.  You seem fixated on that.

The thing us, being big doesn't at all guarantee they're right.

You're the one who thinks you know the market better than Valve does. Sure, being big doesn't guarantee they're right. It does mean, however, that they are not serving the same customer base you are, and hence your anecdotal experiences don't count for much in this discussion. Because while they gather Steam telemetry from millions of computers, you talk to a few tens of people when they come to your shop. 

Doesn't mean the whole idea won't be a dud, but if so, it'll likely be because of the pricing, not necessarily because of any issues inherent with its design. The idea behind the Steam Machine is sound, and if it doesn't end up costing $1200 but more like $700, it'll be a fine toy enjoyed by millions of satisfied customers, who'll never feel the need to go to a PC builder. It might expand the market for PC games into the traditional console turf, as well as make Linux a serious competitor to Windows in the mainstream gaming market. Enthusiasts will dismiss it, same as consoles, as a crappy black box that doesn't have the oomph to drive a 244 FPS 4K monitor or a pro-grade VR headset, but that's not going to bother Valve one bit (although if it can drive Valve's own headset, it might boost VR gaming, too).

Edited 17 hours ago by Dragon1-1

[MAXsenna]

22 minutes ago, kksnowbear said:

Well, sometimes the best answers aren't the easiest, and sometimes there's just no substitute for determined legwork and putting in the time.

Those in some way bothered by the length might simply consider not reading it.

I don't think they're bothered by it. But TIME is a commodity these days. If one is not interested in the details, and they can have their simple question answered by Yes/No, in the first nine posts. They might not read your. 

[kksnowbear]

10 minutes ago, Dragon1-1 said:

You're the one who thinks you know the market better than Valve does.

Never said anything of the sort.  I have a very distinct position on what they're doing, but that doesn't mean I'm trying to claim I know the market better, at all.

10 minutes ago, Dragon1-1 said:

Doesn't mean the whole idea won't be a dud, but if so, it'll be because of the pricing, not necessarily because of any issues inherent with its design.

You have absolutely no way of knowing that.

Edited 17 hours ago by kksnowbear

[kksnowbear]

15 minutes ago, MAXsenna said:

But TIME is a commodity these days

I understanding what you're saying, but I'm not forcing anyone to read anything.

In fact, I can all but assure you that the people I've been able to work with and help is a much smaller number than those who insist on characterizing me as some sort of ...bother

There again, it's the ones I *can* help that I do it for, not the "haters".  TBH If I gave up the moment those people started, there wouldn't be anyone here to help the ones who understand the importance of putting in the time and legwork.

Edited 17 hours ago by kksnowbear

[Aapje]

I checked where this $1200 figure comes from, and it seems to come from Vice who are not even claiming that this is the price. They wrote an article based merely on Linus saying that he heard from Valve that it will be priced more like a PC than a console. Then they start throwing out various prices based purely on speculation.

See: https://www.vice.com/en/article/steam-machine-price-range-may-have-leaked-and-its-not-cheap/

So please stop acting if $1200 is an actual leak. It is not.

[Dragon1-1]

The points made still stand, though. "More like a PC than a console" is exactly the wrong approach to take with this thing. If it's priced like a PC, it will face stiff competition from actual PCs, which bring a lot more to the table.

1 hour ago, kksnowbear said:

You have absolutely no way of knowing that.

Yes, and you have absolutely no way of knowing that it will be a dud. All we know is that, contrary to your claims, physics don't prevent it from working just fine given its stated specs. They could do something dumb, but that's no reason to dismiss the concept. As above, I see the biggest risk is them overpricing it.

Worth noting, there are good reasons to hope it succeeds, since any game natively supported on SteamOS should run on most other Linuxes natively. It's probably the closest thing we'll get to an OEM Linux machine outside China, too.

Anything that lights the fire under MS is a good thing for us DCS users, as well. While it's unlikely DCS will run natively on Linux in near future, an incentive for MS to debloat Windows a bit would nonetheless be appreciated.

Edited 16 hours ago by Dragon1-1

[kksnowbear]

1 hour ago, Dragon1-1 said:

Yes, and you have absolutely no way of knowing that it will be a dud.

Not entirely an accurate characterization.  I have a lot of background in professional capacities, knowing exactly what will and won't fail, much of it in applications with practically zero room for error, and actually in the computerized systems industry (as opposed to whatever other completely unrelated discipline).  Sure, a plumber is also a professional...but I'm not hiring a doctor to build computers, or an insurance agent as a lawyer.  I'll consult a nutritionist when I need help with...nutrition.

I'm not going into it further, because I'm tired of being attacked for my (perfectly qualified) observations.

Edited 14 hours ago by kksnowbear

[Aapje]

3 hours ago, kksnowbear said:

There are people who understand that certain things inherently contain weaknesses. They don't need to know a price to identify problematic areas in a design.  Engineers do this all the time, when price isn't even part of the discussion.  Some things are obviously bad ideas, irrespective of cost. If the design of a bridge won't bear the load, an engineer had better know it, never mind the cost.

Yet you don't seem to understand that there are different use cases.

Mature, useful advice explains the use cases that the device is not good for, but also what it is good for. You completely fail at that latter part.

Fact is that 10's of millions of gaming laptops are sold each year, sales of those are increasing and people very often buy another gaming laptop, so they often seem to have no regrets. Yet a gaming laptop has similar heat/wattage limitations to a small factor PC. Furthermore, people also buy consoles, also often small designs with little upgradability.

Actually purely bad ideas are the old Alienware desktops, which had horrible thermal engineering and lots of custom components hampering upgrades. And these were full-sized desktops, so you didn't even gain portability or space efficiency, which are actual benefits of small form factor PC, gaming laptops and consoles.

But a trade-off that you personally don't like is not automatically a bad idea and it is also not necessarily something that the market will reject.

 

[kksnowbear]

27 minutes ago, Aapje said:

a trade-off that you personally don't like is not automatically

I didn't say it was "automatically" anything.  But it is, in my estimation, a design with unnecessary potential for reliability weaknesses.

Millions of people buying underpowered "gaming" laptops doesn't prove they're a good idea.  It simply shows people will buy anything that is marketed properly, that's it.  They're good for traveling people who like to game, or similar situations.  Performance is sacrificed for the sake of portability.  That's fine, just call it what it is.

It's not about what I don't like personally.  You consistently try to bring "personal" into a discussion.  It's not personal, it's an informed and qualified professional observation. 

Your opinion about my assessments is inaccurate.  I already identified my position on applications these SFF devices were good for.  Just because you don't agree doesn't mean I "completely fail" anything.  Plenty of people trust my professional observation, and pay for it, too.  Is it possible for you to have a discussion without the personal attacks?

Edited 15 hours ago by kksnowbear

[kksnowbear]

Anyhow, have a great night fellas.  I have better stuff to do.  Buy all the underpowered miniPCs ya want.  Tell others, too.

That way, when they all finally decide to step up to real gaming computers, there will be more clients for builders like me

Edited 15 hours ago by kksnowbear