Sony's Deal with Microsoft Blindsided Its Own PlayStation Team
bloomberg.com
bloomberg.com
Uh, no. This shows that the developers have absolutely no control, say, or ownership over their work. Things like this happen in companies (or divisions within a company) right before their best people jump ship because they are so poorly respected they are literally not given the time of day by management.
I'm reminded of Ray Kassar's comment about programmers being at the same level of someone who puts the game boxes together.
Developer here. What role would you see developers playing? This is more strategic than technical, so I wouldn't expect much developer input at this stage.
If management asked and the developers didn't like it, people would say "why the opinion charade?". So the only options are don't partner or get a strong (whatever that means) majority vote from the developers before doing anything. Both seem like bad options to me.
Software is a new form of literacy. Literacy makes all the difference - you don't hire illiterates to run companies, they might be clever, politically astute, but the decisions they make will be wrong from the important perspective of "does this make sense in a literate world".
Imagine you were hiring people for a trading company in the 16th Century. If the top people could not read or write, would you expect that trading company to perform as well as a competitor that had hired its board, its execs, as those who were able to red and write. Who expected to read the ships manifest?
Companies that ignore the developers are companies that are ignoring what literate people think is sensible. Now sometimes that will be the "right" thing to do, but increasingly *when illiterates and literates clash over policy, the literate side is more likely to be the right one to take"
Just my 2 cents
The typical C-level dictatorship provides neither of those (besides just being a plain miserable environment to work in).
How are they going to make a Google Stadia style service that requires custom hardware units in data centers around the world with the lowest possible latency to the end user? And how are they going to do it quickly?
I guarantee Microsoft has a competitor to stadia ready to go, they have the data centers, they have the development expertise and they've been pushing xbox apps onto every platform that will take them.
If Microsoft will give them a good long term rate on the custom azure instances that the xbox team will be using to host the games then that's probably a whole lot better than Sony trying to roll their own.
I mean....they already do, so in that sense they are ahead of Stadia. Playstation Now is a real service that exists and which you can use already, and which required custom hardware in datacentres around the world. Now we can argue whether it's as good as Stadia promises to be, but that's purely theoretical at this point - one product is an actual commercial thing you can get, the other one exists in beta stages.
And I am simply transferring the same intuition about this to software.
If "Google SRE is what you get when you ask a developer to design an ops team" then what do you get when you ask a developer to design a whole corporation. Or a government ministry?
People who can code are not fundamentally superior to everyone else as you say, but they offer different ways to organise as much as people who 300 years ago were literate were not fundamentally superior to the rest of us, but offered different ways to organise.
So here is my simplest idea - as pretty much everything is being eaten by software, then project management should flow entirely from code. How is Project X doing? Ask the codebase.
[#] Bicycles of the mind if you wish. But something important is happening - its beyond telegraphs or telephones in enabling communication between humans.
Software is, ultimately, a set of logical instructions. Other people who understand those instructions are literate in that domain. You seem to be taking a single application of literacy (software) to be the entire scope of literacy writ large. It isn't clear to me why you think that, especially when no programming language is capable of "encoding" the range of human thought that, say, English is.
Please respond in the programming language of your choice.
I don't think that's fair. Software if anything is a means of modelling and transmission of that model - being able to create and manipulate models is what we try to do with language and written word - but the compiler we are targeting is on other people's heads.
so it's a bit hit and miss - we spend a lot of effort making us all share the same compilers, from schools to legal judgements.
but with software the compiler is simpler and so the models can be more transparent and that Inthinknis the big difference.
No I cannot express my model in english as well as I can in any program- I am not trying to express the whole range of human thought in software - but what we can do is and will be so valuable I can only compare it to the shift in value we got from mass literacy
But knowledge of the software world does not (dis)qualify one for business leadership, or to be a heart surgeon, or a janitor, or any number of fields. Your original post implied that knowledge of software, the "new literacy" is a prerequisite for any moderately advanced task, which is nonsense.
I disagree - I think software is more horizontal - it allows great improvement of any form of knowledge work, and is much more comparable to "learning to read" (or perhaps sits somewhere in a hierarchy of horizontal skills between "reading" and "statistics")
Do you consult with the admin staff to check they're okay with your code?
In large businesses (in particular) you have to have delineation of roles - everything can't be democratic. Just as (I assume) you expect your seniors to trust you to code competently, so you should trust other people to handle strategy competently. You might not like the end choice… but that’s usually because via not having been in the discussion, you lack key information as to why the other smart people in your company came to that end choice.
(One thing I have learnt: everybody wants to be involved in strategy, because it’s fun, and because they think they can. I see this in my own industry all the time. People should have the discipline to stick to the scope of their role.)
One of the things that came with increased literacy was increased middle class and increasing widening of the suffrage.
I might have something to say about the idea from a technical perspective, if asked, and if that aspect of the platform were my purview, and I suspect they discussed it with plenty of knowledgeable people at some level.
I'm sure some people who were developing their in-house cloud solution feel like they've been blindsided, but if they'd succeeded, they wouldn't be in this position. The article says they've been struggling with it. They likely had a chance and missed benchmarks one time too many.
They're likely better off using a major cloud provider anyway. They've already solved the big problems.
Are you suggesting that developers deserve some kind of vote? They're a cog in the machine. Sony isn't a democracy or a worker's collective. As a developer myself, I'm not trying to diminish the role. I like to think I can provide a valuable perspective and that I'd be consulted about my small part. I'm certain that's exactly what Sony did, because they are a successful, well-run company that's been doing this longer than I've been alive.
Beyond that, no, the developers shouldn't have much say. Their role is to help determine what's required to achieve the objectives set by management. Good management will consult with them, set goals, and change tack if the objectives aren't met. If they still can't achieve the objectives in a timely and cost-effective way, management will replace them and go with a vendor. This is all right and good.
Of course not.
Their role is to help determine what's required to achieve the objectives set by management. Good management will consult with them, set goals...
This. Consider yourself lucky that you've never worked someplace that starts off with "These goals are aggressive, but in my mind achievable" and then proceeds to give you objectives that are a wildly optimistic extrapolation of a of a previously achieved timetable.
This isn't just developers, either, but that's the context here.
Any good executive consults with their subordinates to create realistic objectives. This is so fundamental that it's implied.
...
I'm certain that's exactly what Sony did, because they are a successful, well-run company that's been doing this longer than I've been alive.
If you're right them I'm onboard.
~130x as many people bought PlayStation 4 than subscribe to PlayStation now at just 20$ per month. While they might be extremely profitable for Sony, AAA games require huge customer bases.
However, most people are relatively light and irregular users. For a $5 a month service, that may be OK. But as the price goes up it gets less and less attractive for anyone who isn't a heavy user.
I don't mean all software developers, but it's a common refrain on this forum.
The reason is that they're temporarily in demand. Ask the people who were around during the dot-com bust how much the average programmer was valued then.
As someone who has worked in a number of places and ended up programming... I find this absurd.
Lots of programers have no idea how good they have it / influence they have.
Just because someone could make a call without their input doesn't mean they are like manual labor.
All this report says is that rank and file had to be calmed down after the reveal (fair enough). That doesn't mean that technical input wasn't asked or, or even listened to.
Like, I can totally see something like HQ asking "What would it take for you to grow X, achieve Y, by Z", and then PlayStation unit saying something like "M people, N money, O risk". And then HQ was like "oh, MS can do better than that".
The PS4 and Xbox One use very similar hardware. In fact, other than DRM, neither is that different from a gaming PC.
It seems you'd have no trouble running PS4, PS5, Xbox One or Xbox Two games on the same hardware box that would be racked in an Azure data center.
For that matter, Microsoft has Azure data centers all over the place and very good networking for those data centers. A few years back I noticed that the nearest Azure data center had 35 ms latency to me compared to AMZN's us-east-1 which had 50 ms. (Maybe that is why AMZN built us-east-2 across the street!)
I don't see any reason why Sony should build a whole bunch of data centers and struggle with second-best networking when MSFT has already made the investment and is continuing to make the investment in a great data center product.
Big companies rarely collaborate well on wide-ranging things unless the companies share board members/heritage.
This deal was clearly spearheaded by the Azure sales team, and vague promises of collaborating on future services were probably thrown in to sweeten the deal, but I doubt will go anywhere.
https://www.microsoft.com/en-us/Investor/earnings/FY-2019-Q3...
Our Microsoft rep recommended we consider deprecating our SAP instance with Dynamics CRM as we could get more discounts and support hours as part of a better valued package and our CIO in that meeting laughed in her face.
Azure might be just as good as their competitors in 2019 or even better but your question reminded me of way back when it was a piece of shit and our clueless CIO was mostly bought in because Gardner could have told him to fellate a horse, and he would have obliged.
2018/19 that's still the only kind of CIO that would choose Azure if the choice is otherwise unconstrained. Most Azure offerings are somewhere between Alpha/Beta stage of maturity with glaring issues, only cursory (often outdated/outright wrong) documentation, zero community around them.
The only way Azure gets picked is if the decision maker is an idiot guided by the likes of Gartner, or there are other factors at play. Other factors usually being the Azure sales teams leveraging existing MS mindshare and pandering to entirely non-technical decision makers with co-investment or support, the sum value of which is still far less than the cost of choosing Azure.
Not saying that people don't complain about them or that there aren't better things out there, just that they are good enough
> I am thinking of one high-profile Internet startup in San Francisco right now that is extremely promising, has great technology and a unique offering, that did two big deals early with high-profile big company partners, and has become completely hamstrung in its ability to execute on its core business as a result
Personally, I see it as two entrenched players seeing a potential newcomer (i.e. Google Stadia) and realizing it would be better to work together and keep the duopoly going than let a newcomer tear away both their market shares.[1]
It's always better to pick the devil you know, as they say.
[1] https://www.polygon.com/2019/5/16/18628245/sony-microsoft-co...
After Sony, Microsoft, and Nintendo I'd imagine the largest competitor/concern would be either Apple or Oculus (with their new Quest console), Google perhaps next, but that has nothing to do with this alliance. Google cannot even encourage games on a tablet or Chromebook, let alone a TV.
Apple are a slightly bigger threat because the Apple ecosystem already supports larger and richer games, including on tablets and TVs.
[0] https://en.wikipedia.org/wiki/Cloud_gaming#Cloud_gaming_serv...)
You can't really refute the actions from these companies themselves; they speak for themselves.
MSFT probably agreed to go in giving Sony good terms because they are afraid of Stadia. Not because it’s a threat to their gaming business but because it’s a threat to Azure. Heavy Stadia use means GCP has a high use tenant that pays for their use which they can use to make the rest of their offerings cheaper without losing profit. And that would hurt Azure, which is gonna be huge if MS is gonna live up to its trillion dollar valuation.
IOW they are worried that Google Stadia will be to GCP what Amazon retail is to AWS.
IMHO the deal is mostly coming from the fact that managing data centers is a mess, and MSFT managed to convince Sony they could do it better and for less money.
google can't into long running services anymore, their company culture has become toxic to it
From what I understand, even Apple is starting to build more of its services infrastructure in the cloud, although those agreements aren't being widely broadcasted.
With that view, it should be obvious that a hardware-focused company like Sony simply could not scale the internal teams and infrastructure required to support high scale software services like on-demand gaming. They were going to have pick a cloud infra provider, and with every cloud provider also making their own foray into gaming, they had to make frenemies with someone.
If you have a good on premise service provider and reasonable compute/scaling requirements, staying on prem can actually save you a ton of money. It’s very situation specific.
That being said, for companies starting out, I agree completely with you that starting in the cloud is the best strategy.
I think they qualify - Sony need rackmounted PS3 and PS4 hardware, since the games in question won't run on standard AWS/Azure GPU instances.
The PS4 is built with a custom Ryzen-based APU. Because Ryzen is x86_64-based, it's likely possible that they could also make machines that are architecturally compatible- they could probably get away with a compute box filled with similar, if not better, graphics chips / cards, use an EPYC processor with all its cores and PCIE lanes to connect everything up, and then virtualize it.
For the PS3, however, they would definitely need something very custom- their Cell processors are very different from anything else on the market, but they also already have a service for that, but I have no idea how those work- they obviously have something, though.
This is the kind of nonsense the so called AAA companies are trying to force on their customers: https://mobile.twitter.com/anthemgame/status/104320071518584...
Consoles are exceedingly cheap nowadays. Even on the high end, it's a $400 (PS4P) cost spread over 4 or more years, which can work out to as low as $10/month. I haven't seen any "production" cloud streaming service that cheap; some can be cheaper if you don't play a lot, as they bill per hour of active time. Even Xbox Live Gold/PS+ isn't cheaper than that, and that's a FAR simpler service to run versus cloud streaming.
Services like Stadia or xCloud will cater well to the casual gaming crowd; 10 hours or less per month, not seriously invested. That's not where the money is at (in consoles, its DEFINITELY where the money is at in mobile). That's the flaw of all of these services; there's no way they can be offered at a cost competitive rate to running code inside the house, and that's not even considering internet speeds. The funny thing is, at least in my experience, it's the "serious" gamers who have the best internet and thus would be most eligible for these services, yet would also be most likely to want the best quality experience (audio, visual, input lag).
These are the gamers that would gladly pay this month to play the newest games on a virtual console, rather than pay up-front.
But Stadia actually interests me for a different reason: I would love to be able to play my games on different devices at different times, especially away from my house. Phone, tablet, laptop, home computer, and Android TV box... I would play those games on those devices at different times.
If my PS4 could stream to them all reliably, especially when I'm at my in-laws house and my PS4 is at home, I'd be doing it.
Stadia is basically promising that.
Microsoft did announce that "Xbox subscription" service at some point. I'd challenge you to start at store.microsoft.com and find it; I can't. They probably still offer it somewhere, but its not a priority for their sales, and I seriously doubt it represents even a small portion of their console ownership. It was an experiment.
The issue is, everyone is focused so much on the advantages of Stadia-like services we forget all the INHERENT downsides. There will be downsides. Of course it would be great to play any game on any device at low latency for low cost. This WILL NOT be even CLOSE to any future we will experience. Here's what will happen:
- Gaming exclusivity is still a problem and only getting worse (see everything going on with the Epic Games Store; even on a single platform like PC there's still extreme fragmentation). (Very) Arguably the best games of this generation are on PS4. Sony is killing it with their first party studios. Stadia and xCloud won't get these; so we'll have to adopt multiple streaming services, with varying degrees of cost, quality, and end-device support. Note: This is identical to our situation today! Fragmentation has always existed in gaming, and it always will; streaming will make it worse, not better.
- End-Device support and quality of experience are tightly linked. Google may have the engineering chops and scale to get it right; they know the browser, they have deep networking and latency expertise, and they have DCs all around the world. But even that may not be enough. Game streaming would most deeply improve the gaming experience on the literal devices that are the least suited to actually offer an enjoyable experience (tablets, phones, wireless devices in exotic locations). The funny thing is, game streaming would most impact the ability to offer great graphics on these devices which traditionally couldn't do it, but gamers CARE WAY MORE about latency, they just haven't ever had to worry about it because we've always had local processing. Check out some of the Stadia first impressions [1]; they're hopeful, but also it's very clear that even in the best situations this technology degrades A/V performance substantially while still having issues with latency.
- No one is discussing cost. Google won't touch the subject. Even if they get everything else perfect, even if the experience is fantastic and identical to local processing (it won't be), I very strongly believe this is the single point that will kill these services. This is why there are other comments in this post which say "the only people who care about this are the corporations, gamers don't actually care". There is ZERO CHEAPER WAY to deliver processing to a consumer than shipping silicon to their house. We've already seen this with cloud transitions at large companies; it ends up being more expensive, but they don't have to manage large operations teams. Well, I don't manage anything with my PS4; I bought it and it sits on my tv stand, so that's out of the question. But the moment you introduce DCs and global networks, now Google/MS have to pay for dozens of engineers to support this thing that was previously just dark silicon in my living room. It CANT be cheaper. At best, they may charge a flat fee then hope the number of casual gamers (<10 hours per month) outweigh the number of serious gamers. At worse, and most likely, they'll support it with advertisements, micro transactions, or degraded quality. Actually, in Google's case they'd probably just shut it down.
- Speaking of cost: game studios are fucked. How do they get paid in this new subscription-based system where there's a dozen players taking their 5% tax at every turn? There's something beautiful about a simple 70/30 revenue share for a product sold on Steam, even if the tax is high. With subscription game streaming, maybe it's based on time spent in game? Look forward to more games that inject insane filler content just to make sure you HAVE to spend 40 hours in it. Look forward to more microtransactions and DLC that aren't included in the subscription. This is the music industry all over again; Spotify "saved" the music industry from a piracy epidemic, but today's reality of royalty payments is still pretty shitty. The gaming industry doesn't have a widespread piracy problem; the game subscription debate is just technologists (corporations AND geeks) inventing a problem where there isn't one, while the artists, who already don't make enough, who work at companies that are seriously struggling (outside of the very few "winners" like Epic or Valve), silently raise their hands and sheepishly say "team, what about us?"
- Let's just ignore Nintendo, right? Creator of the best games in history, the only major player seriously committed to creating experiences that span all age ranges, including some of the best games this generation, and they have shown zero interest in game streaming. Good on them.
Screw cloud game streaming, screw game subscriptions. I'm fed up with the technology giants constantly seeking out their 10% "service tithe" in every possible vertical they can find, while fucking over consumers and artists in the process. Microsoft and Sony are fine; they recognize that they need investment in this area, but I truly believe they actually love and support gaming, gamers, and game developers and don't want to see a world where we get screwed over. As always, its Google who should be ashamed; they're forcing the hand of the industry to invest in this, but we all know they won't succeed (they can't even build a messaging app, they had to BUY a video streaming platform and they've proceeded to really screw it up, they expect to create a compelling set of first party game studios with no experience?) Google doesn't actually care about anything except Eyes On Their Products, and they'll do anything to stay relevant.
What technology were you playing?
LAN parties are great, but you don't need super-portable if you're packing for an all-day/overnight trip anyway. A computer that fits in a duffel bag is fine, and plenty of "luggable" gaming laptops qualify.
With current geo-synchronous satellites, correct. With the upcoming low earth orbit internet satellite constellations of which there are at least 3 announced and the spacex starlink constellation is literally launching their first batch of 60 satellites this week, this should be doable.
Also, lots of people prefer digital copies of games. Just look at the digital sales vs physical sales for Nintendo games which are generally available for both. I presume this is the case for other systems as well.
I'm interested, assuming a reasonable price and performance level. Not having to have gaming PC plus being able to easily play a bunch of new games for a few hours without shelling out $60+ each are both things that sound interesting to me.
Then again I think the last physical game I bought was Europa Universalis III a decade ago, so physical copies haven't really mattered to me for a long time.
The gaming monetisation sphere has come a long way since EU 3
The closest we got (if I remember correctly) was Darksiders 2 launching on OnLive at the same time as PC, but it was also a rental game rather than being part of their Netflix-style subscription PlayPack.
1. They'll likely fail in their gaming effort with Sony still being in strong position in market.
2. They have a competitive cloud with bright engineers that can solve infrastructure problems better than Sony.