Interview with Jason Ronald of Xbox Series X: "I do not think that the consoles will disappear soon"


'The Xbox Series X is almost around the corner. Microsoft has revealed a good part of the advantages that we can expect from it, but there are still many unknowns to solve, and today we have had the opportunity to answer some of them.

We have done so thanks to the interview that Microsoft has given us with Jason Ronald, Director of Program Management and Xbox Series X. With him we have been able to speak a lot and well about a console that promises power and flexibility.'

Read Full Story >>
The story is too old to be commented.
Marquinho92d ago (Edited 92d ago )

"We could have used forced clocks, we could have used variable clock rates: the reality is that it makes it harder for developers to optimize their games even though it would have allowed us to boast of higher TFLOPS than we already had, for example. But you know, that it's not the important thing. The important thing is the gaming experiences that developers can build. "

Exactly what I've been saying. Variable clocks are to make the system look more powerful than how it effectively will in real (heavy loaded) gameplay.

Predictable performance is always better than variable (aka, unpredictable) Add to that massive horse power, smaller form factor which equals more efficient cooling and it males the Series X the biggest technological achievement in consoles... probably ever.

Pricey92d ago (Edited 92d ago )

That's not what he said though. He said he thinks its extra work for the developers. He did not say that it won't have it's stated performance.

Also how does the smaller form factor indicate a more efficient cooling solution. Firstly if you consider volume I don't imagine the form factor difference is massive. Secondly one console is running at a much clocks so it will likely need more space for cooling, that doesn't make it less efficient. Thirdly you make all these assumptions without knowing anything about said consoles cooling solution.

--Onilink--91d ago (Edited 91d ago )

Based on the dimensions estimated so far, I would still bet volume wise the difference will be significant, in particular on the disc drive version.
That aside, everything else he said was nonsense and we dont know anything officially about cooling on the PS5, but if its using a vapor chamber like it was implied on the patents, then its definitely a great start

xenz92d ago

Smaller formfactor does not equal better cooling. The more space you leave for the heat to evaporate the better.

Marquinho92d ago (Edited 92d ago )

^ ^

If it could sustain those variable - peak - clocks, then it wouldn't be an issue for developers. He doesn't have to redound on the obvious. That's the very point of "peak" clocks... they're not stable, they're hard to predict, hence, involves extra work for developers unless they just stick to the maximum sustained clocks the machine can offer. In PS5's case, it's said to be 9.3TFlops whereas on Series X it's (confirmed) 12TFlops at your disposal for 2 seconds or 2 hours non stop. That's the beauty behind this man's team work.

As for cooling, you need a state of the art cooling in order to put 12TFlop in such a small case. That's what makes Series X (and One X) such an achievement. Making the case bigger to avoid overheating isn't really something to be praised about.

Pricey92d ago (Edited 92d ago )

Its harder to fill the extra CUs with useful work that make up those 12tf in the Xbox Series X than the PS5. A point that Mark Cerny made in his presentation. Its easier for a developer to have slightly fewer CUs and higher clocks, there is a trade off though which is the added complexity of the dynamic clocks.

Marquinho91d ago (Edited 91d ago )


More CUs = More RT Cores. That won't be a disadvantage just like multicore wasn't back in the 360 days. I think you're trying to hard to find a power disadvantage on the Series X, but it just isn't there.

Btw... Cerny's expo felt like a handful of reasons (excuses) as of why they couldn't match the power of the already announced Series X. That "variable" clock figure is controversial because it just doesn't reflect the real sustained power of the console. It was Sony's way to avoid staying at 1 digit (9.3) and market their console as 10.3TFlop... when it isn't.

Keep in mind Sony' have done "last minute" changes to match their rivals before, like the 8GB of RAM on the PS4 (it was going to be 4GB) or the inclusion of "Tilt" on the Sixaxis, to somehow gimmick the Wiimote.

Pricey91d ago

That last minute RAM change was said to be in response to developer feedback. I expect the series X to be better at ray tracing. I expect the PS5 to produce better object/world detail since it can stream data in faster due to its faster SSD.

Your making assumptions you can back up again, and it just looks, to me at least; like you need one console to be better than the other a little too much. It's not healthy.

The dynamic clock approach is only controversial because it's not conventional. You don't know yet how often the console will dip in performance. Until proven otherwise I like to give the guy who built it the benefit of the doubt. He said in extreme (limited) circumstances the clocks may drop. They will only need to drop a very small amount too, we are parking single digit percent. As part of the tooling developers will be able to optimise their game so that the sort of spikes that could lead to these drops are even less likely to happen. I intend to believe this is until proven otherwise, and if what's stated is true you may not even notice when the drops occur.

Marquinho91d ago (Edited 91d ago )


=) ... We all know the RAM increase was in response to MS offering, which leaked before E3. That's how the market works though. I'm not against that.

What I am against, is misleading advertisement. I find it fair enough that you want to wait for benchmarks of multiplatforms to tell the actual difference. My bet is that it will show a difference way above the 17% many PS fans are expecting due to Sony's controversial strategy for selling its console's power.

As for PS5 producing better detail... that's not going to happen. SSD wont let PS5 render more than what the GPU is capable of moving. XSX btw, will also use its SSD as a RAM extension and it looks like there's something this guy has under the sleeve referring to that ("it goes way beyond numbers we can or cannot share") Since this is totally new and we don't know to what extent the Series X Velocity technology will improve the actual I/O capability on the system (according to DF it will boost it beyond 2x-3x) I will give it a "wait and see" status. (For comparison, this is also new, but not controversial, just intriguing)

Then again, make no mistake. I'm absolutely getting both machines. I'm a huge fan of God of War, Bloodborne and the first TLOU. I respect Guerrilla a lot too... have high hopes for Horizon 2. I'm just trying to be as accurate as possible in terms of both system's capabilities.

PS. Good debate btw, thumbs up 👍

Pricey91d ago

I also intend to get both, our opinions are probably more aligned than not. I just happen to think the two systems are closer in terms of capability than you do and I'm also equally impressed by what both have achieved in terms of architecture.

I stand by my assertion on level of detail though. IO remains a major bottleneck where gaming systems are concerned. The bottle neck is reduced by adding a SSD but it's not removed. It's still slower than RAM. The reason pop in occurs so often in the current generation is because of IO not the GPU. Developers try to solve this by having more data resident in RAM than they might actually need. This wastes RAM. If you have a super fast SSD you can swap out RAM more quickly and therefore can fill it will less data but of a higher fidelity.

+ Show (3) more repliesLast reply 91d ago
92d ago
Hakuoro92d ago

I fail to see how it makes any extra work, this is a feature that both CPU and GPU's have had for 20 years. The boost happens as a function of the CPU/GPU and it's not something that has to be coded for in development.

But w/e it wouldn't be a Microsoft interview if they didn't downplay Sony.

Atom66691d ago

And it wouldn't be a Microsoft article if the Sony fans don't come running in.

Gaming4Life198191d ago

But w/e it wouldn't be a Microsoft interview if they didn't downplay Sony.

Sony has consistently thrown jabs at ms this entire generation so if ms was throwing shade who cares. I dont even think he was downplaying sony, i think he was stating why they didnt go that route.

AngelicIceDiamond91d ago

Downplay? Cerny gave his reasons for lesser CU's and nobody saw it as foul play. MS is simply explaining why they didn't go with boost modes but some hoe that's foul play? These companies have different philosophy's when building these systems.

It's not healthy cheer leading for a corporation like that.

RamRod8891d ago

Dude it's always console wars with you. The guy was asked a question about why he went one direction vs another and he answered it. He didn't say anything about Sony or throw any shade.

+ Show (1) more replyLast reply 91d ago
umair_s5192d ago

I don't think Microsoft strategy is only just PS5 vs Series X. Xbox gaming is going to be much broader than a single device. The Xbox ecosystem is going to work on a range is devices thanks to project Xcloud.

Kribwalker91d ago

that’s the whole reason they changed their metrics to users. Everyone was saying it was because they were embarrassed with sales numbers, meanwhile they’ve put their games on pc to broaden their market (which has worked as they have generated more revenue selling less consoles with pc releases then they were during the 360 days), they’ve added sub services and shortly they’ll have hardware free with xcloud.

they aren’t only playing the console game, which is why consoles sold =\= success for them but Users does

crazyCoconuts92d ago (Edited 92d ago )

I agree with everything but your last paragraph. I'd rather have a more powerful gpu running at reasonable and steady clocks. I'm sure Sony made their choice to make the thing affordable. Given that I'm assuming that the smaller ps5 GPU + the more advanced/costly cooling system they're probably deploying to counteract the heat must still be cheaper than, say, the Xbox gpu.
The rest of what you're saying, I don't know... Variable has trade offs not defacto inferior, form factor not sure it's smaller in volume than ps5, but it doesn't need as nutty of a cooling solution because it's running at lower clocks. Still, I'd rather have the XSX GPU in the upcoming PS5 no doubt

Marquinho92d ago (Edited 92d ago )

Keep in mind the "cooling" solution for the Series X doesn't only involve the vapor chamber, form factor and physical design. They actually split the motherboard in two in order to achieve their power/size/cooling goals.

According to Arstechnica, PS5 will be the biggest console ever made with ~550 cubic inches (That's even bigger than the colosal original Xbox... yes the one from 2001) while the XSX will be 418 c.i.

That gives you ~30% of more 'sustained' power in a 30% smaller case. If that isn't a hell of an achievement, I don't know what it is.

Pricey92d ago

The PS5 is only an estimate since no one has actually held one yet. And the Series X is 418 cubic inches.
But this does not tell you the cooling solution is better, as i and others have already stated.

Again, you have no idea what the PS5 cooling solution looks like. You seem to think just because the series x has a nice cooling solution the other console is automatically inferior.

Hardware architecture != Cooling solution. If the PS5 is bigger that doesn't mean the cooling solution suddenly inferior, it might mean it runs hotter and needs more cooling.

crazyCoconuts92d ago

I think you have good points. It would be 25% smaller if those numbers are true, but you're right, more power in a smaller package is impressive.

Marquinho92d ago (Edited 92d ago )


It's not that complicated. Just think about this:

~20-30% of more power on a ~30% smaller case. (XSX vs. PS5)

Which one's a bigger achievement?

Pricey92d ago (Edited 92d ago )


One has an SSD which has at least twice the bandwidth 🤔. Note many of the fastest SSDs on the market come with custom cooling. Some even have active cooling.

Both are fantastic. They just took different approaches.

Oh and I think your being a little liberal with the power percentages to the point of being disingenuous. You don't have to reach that much to convince me that the series X GPU has more compute power.

+ Show (2) more repliesLast reply 92d ago
RazzerRedux92d ago

Well...the thing is that what he is saying it totally at odds with with Cerny said so, once again, guess it is going to come down to actual benchmarks to really see the difference.

Kribwalker91d ago

What’s better

The ability to consistently access more power with no loss


having to throttle one down to bring the other up?

RazzerRedux91d ago

XSX varies electrical power while maintaining a constant frequency. PS5 varies frequency while maintaining electrical power. There is a loss either way as thermal limits exist in both scenarios. These principles apply to every system.

IRetrouk91d ago

One doesnt have to be throttled down to bring the other up though, both the CPU and GPU can be ran at their highest clocks at the same time, this has been stated by cerny himself.

The variability is based on what's going on in the game, if the devs need the chips to run at their max frequency the system can do that, but as cerny and other devs have explained, games dont need to be constantly running the system maxed out, it's a waste, cernys design seems to combat that.

thebutcher0990d ago (Edited 90d ago )

If one does not need to be throttled down for the other to hit max frequency then why are they using AMD's smart shift tech? If both the CPU & GPU both have enough power to hit max frequency they would not need a technology that distributes power where it is needed.

IRetrouk90d ago (Edited 90d ago )

Smartshift allows both the cpu and gpu to share a power budget, if either the gpu or cpu are not fully being used the remaining power can be transferred between them. Nothing to do with being able to run at full clocks or not.

thebutcher0990d ago (Edited 90d ago )

Let me clarify, when I talk about running at max frequencies I am talking about running at max on a constant basis (we do not know the potential TDP impact of running at those clocks for prolonged periods of time). I know what Smartshift is but the question remains, if there is enough power being supplied to both the CPU & GPU for each to run at max frequency, consistently, then why the need to "shift" power anywhere?

Edit: I think Sony to clarify a few things where there is still some ambiguity such as the scenarios where, most likely, the CPU will be clocked lower for the GPU to run at max freq (in cases where this is needed over longer periods of time) and exactly what the reduction of power on the CPU would need to be to achieve this.

RazzerRedux90d ago

"if there is enough power being supplied to both the CPU & GPU for each to run at max frequency, consistently then why the need to "shift" power anywhere?"

In that scenario there is nothing to "shift". Both are being fully utilized as IRetrouk said. SmartShift comes into play when the CPU or GPU are not being fully utilized. The power budget allows resources to be diverted. On XSX, for example, if the CPU is being under-utilized in a game while the GPU is maxed out, the CPU cannot help and the system overall loses efficiency.

thebutcher0990d ago

If there is enough power for both to run at a constant max frequency you don't need technology to distribute power between the CPU/GPU. If the GPU is running at max frequency and the CPU is not, the extra, unused power from the CPU will not make the GPU anymore efficient if it is already running at max.

RazzerRedux90d ago

"If the GPU is running at max frequency and the CPU is not, the extra, unused power from the CPU will not make the GPU anymore efficient if it is already running at max."

Right, but this is about the shared power budget when one component ISN'T at max power. The problem is if the GPU is at max and its thermals are rising due to the power consumption then it may have to downclock. But since power is shared then the shift can alleviate that problem and allow the GPU to continue running at max power. This is why Cerny referred to both CPU and GPU are running at max as the "worst case scenario".

"Basically, if you were playing a game and the CPU is sitting at 3.5GHz full bore at 35% utilization (just an example) and the GPU is hitting 100% utilization but not reaching full speed *because of thermal/power constraints on the entire system* it can dynamically pull back the CPU and increase the GPU available power resulting in overall increase in the experience and performance."

thebutcher0990d ago

None of my points have disputed that. My point was simply that both the CPU & GPU cannot run at their max clocks consistently and in parallel....I was replying to iRetro's comment of... "One doesnt have to be throttled down to bring the other up though, both the CPU and GPU can be ran at their highest clocks at the same time".

RazzerRedux90d ago

"My point was simply that both the CPU & GPU cannot run at their max clocks consistently and in parallel"

It can as long as it does not exceed thermal limits. That is true of every system. If thermal limits are exceeded then something has to back off. On PS5, it will be the clock. On XSX, it will be power.

thebutcher0990d ago

Quite simply, you are wrong, the PS5 does not look at the thermals to decide when to lower clock speed, it looks at the activities running across each of the processing units. The thermals seem to be pre-determined by the max power budget supplied to the SoC.

RazzerRedux90d ago (Edited 90d ago )

Mark Cerny said explicitly that PS5's frequencies raise until the reaches the capability of the cooling solution. Regardless of the fact that it looks at activities for adjusting frequency this is still a method for keeping preventing collision with thermal limits. So no, I am not wrong.

thebutcher0990d ago

You seem to miss the part where Cerny states "we then tackle the engineering challenge of a cost effective and high performance cooling solution designed for that specific power level" and again if you look at the eurogamer article regarding the same topic..."gives the system on chip a set power budget based on the thermal dissipation of the cooling assembly."

So yes, you are wrong, the thermals are dictated by the maximum power supplied to the SoC. The activities being carried out by each processing unit dictate the amount of power given to each.

Regardless of this you still haven't been able to answer the question...if there is enough power for both the CPU & GPU to consistently run at max frequencies then why would there ever be a need to take power from one to supply the other?

RazzerRedux90d ago

lol....I'll copy and paste the answer you apparently didn't read and then I'm done. Google is your friend and I think you should research this some more as you've not quite got a handle on it.

"Right, but this is about the shared power budget when one component ISN'T at max power. The problem is if the GPU is at max and its thermals are rising due to the power consumption then it may have to downclock. But since power is shared then the shift can alleviate that problem and allow the GPU to continue running at max power. This is why Cerny referred to both CPU and GPU are running at max as the "worst case scenario"."

see ya

thebutcher0990d ago

How amusing that you think I dont have handle on it but you think you do LOL. Your answer does not actually answer the question AT ALL.

If both the CPU & GPU already have enough power to both run at max, then there is no need for power to ever need to be redistributed from one to the other as they would already be running at max capacity. Cerny actually said "we use AMD's Smartshift tech to send any unused power from the cpu to the gpu so it can squeeze out a few more pixels".....again why would the GPU need more power if it already had enough power to run at max capacity.

The fact is I think you need to brush up on your reading comprehension and being able to think in a logical way.

see ya

IRetrouk90d ago

The whole point of the variable frequency is so that the system doesnt need to run full bore all the time, it reads what the CPU and gpu are doing and bases the frequencies on that, no game needs all the power all the time, that's why cerny went the way he did, but you can run both at their sustained clocks, that's from cerny himself,

"Cerny states: “The CPU and GPU each have a power budget, of course, the GPU power budget is the larger of the two. If the CPU doesn’t use its power budget – for example, if it is capped at 3.5GHz – then the unused portion of the budget goes to the GPU. That’s what AMD calls SmartShift. There’s enough power that both CPU and GPU can potentially run at their limits of 3.5GHz and 2.23GHz, it isn’t the case that the developer has to choose to run one of them slower.”"

I understand what you are saying, if it can run full bore all the time, then why not just do that. The answer is simple, because it hasnt been designed that way, and its wasteful, mark does explain his reasons for this too.

thebutcher0989d ago

Sorry mate but I dont think you do understand what I am saying. My point has nothing to do with having the processing units both running at maximum all the time, it is purely a point about the power needed across the SoC.

If I have a GPU which has enough power supplied to run at max frequency, on a constant basis (and therefore running at max capacity), then freeing up unused power from the CPU is not going to benefit the GPU whatsoever as it cannot clock higher than its cap. Therefore if the power budget assigned to the GPU does allow for "max capacity" in terms of performance why would there be a need to (smart)shift unused power from the CPU.

IRetrouk89d ago

Because it has been designed to work like that, do you honestly believe if sony wanted to do a fixed system they couldn't? Ps1, 2,3 and 4 all used fixed clocks, they used variable this time because they see a benefit in it, smartshift is not just used to help control the power budget between CPU and gpu, it also helps tell the system what the chips are going to be doing and bases its budgets on that, it's not all about just running chips at their highest frequency, but I can only give you the answers mark cerny gave everyone else, you arnt questioning us, you are questioning him, he says the ps5 can run both its chips at their highest clocks if devs needs too, but they wont, because again, not how games actually work.

+ Show (16) more repliesLast reply 89d ago
AngelicIceDiamond91d ago

This is interesting as well.

"In fact, he anticipated, "things like the Xbox Velocity Architecture or the API Direct Storage, are things that we plan to bring to the PC: it just happens that it will take some time for the hardware ecosystem to get to this level ."

Is he say that PC will be the lowest common denominstor? Or am I interpreting this wrong? To my understanding VRS, XVA, DS is in house at MS. Which aren't available on PC. So I'm not sure how, say Halo Infinite will run and look on PC.

RazzerRedux91d ago (Edited 91d ago )

VRS is currently available on Nvidia Turing architecture cards such as my RTX 2080 Super. DirectStorage is available on PC via DirectX 12 Ultimate which recently deployed to Windows 10 via GPU drivers. What PC lacks is the hardware compression/decompression as well as the IO bus found on next gen consoles for their SSD implementation.

The question for Halo Infinite is how 343 designed the game to handle non-SSD PCs as they are the lowest common denominator. The Ori developer recently discussed this as a design question for any game that will exist on PC. He explicitly said PS5 first part games will have a freedom in design that others do not.

OB1Biker91d ago (Edited 91d ago )

'In PS5's case, it's said to be 9.3TFlops whereas on Series X it's (confirmed) 12TFlops at your disposal'
Who said that? The Wizard of Oz?
Or some old test at a random frequency that you are happy about? Like the X was not tested at an inferior frequency before too?
OIC another guy much more intelligent than Cerny.

NeoGamer23291d ago

The quote is perfectly accurate.

Your comment is saying that this is better then variable clock speeds. What he is saying is that variable clock speeds are harder to develop for not less powerful.

This a financial implication to game making, not a performance one.

aragon91d ago

Ps4 pro talk with Cerny
One of the features appearing for the first time is the handling of 16-bit variables - it's possible to perform two 16-bit operations at a time instead of one 32-bit operation," he says, confirming what we learned during our visit to VooFoo Studios to check out Mantis Burn Racing. "In other words, at full floats, we have 4.2 teraflops. With half-floats, it's now double that, which is to say, 8.4 teraflops in 16-bit computation. This has the potential to radically increase performance."

They all need to up sell their products Sony and Microsoft.

+ Show (8) more repliesLast reply 89d ago
timotim92d ago

"In fact he was putting in some additional examples: "Maybe that game occasionally loses a few frames in a scene with a shootout, for example, because of the complexity. Well, now we have more than enough performance so that no frames are lost ."

That power is also noticeable in the storage and I / O part: "There is an old game that I went back to and played many years ago. I started it on the Xbox Series X and completely forgot that the game had screens of loading : I was loading the game so fast that the game didn't even have time to show the loading screen. "


timotim92d ago

"Above is one more component: the so-called Sampler Feedback for Streaming (SFS), which some developers describe as a turning point when it comes to streaming the world in which we are immersed and representing a higher visual level. As Ronald indicated, this technology "allows us to load textures and makes the SSD drive" act as a multiplier of physical memory that adds to the memory that the machine itself has. "

Comparisons to the performance of the PS5's SSD seemed to put the Xbox Series X at a disadvantage: Based on those numbers, the transfer rates of the Sony console's storage system will basically double those achieved on Microsoft's, but for Ronald the story was different. " Things go beyond the numbers that we may or may not share ."

Sampler Feedback sounds like a big deal and I'm loving it.

crazyCoconuts92d ago

Ive tried to read up on the claims about the new compression algos and the Velocity architecture, and it's sufficiently over my head to not be able to know one way or another. Can the engineers close the gap of an SSD that's specced out to be twice as fast? I'd guess no, but we'll see as soon as we see the same third party titles loading up on these systems. If MS manages to match Sony's load times on games that would be an engineering feat for the books and I would tip my hat to Microsoft big time. What's your prediction?

timotim92d ago

Prediction? I dont have one. However, based off the technological advances in 1X, their heritage in dealing with hardware and software working together on the PC side of things with DX and such and the amount of time Phil Spencer has given them to produce this custom architecture...I'm inclined to believe that they know what they're doing.

crazyCoconuts92d ago

Of course, all these guys know what they're doing.

rainslacker91d ago

They both have hardware compression/decompression to speed up the bandwidth. They use different formats, but they're both fairly comparable to one another. So, there is no gain for either side on this front. But, compression does mean that data can be moved faster, and hardware decompression means that it's fast enough it shouldn't cause added transfer time.

Compression is also never a set ratio, so in some things, there will be high compression, in others, there may be close to none. Compression is a very inconsistent way to measure potential through put.

Anyone that acts like compression is going to make up the difference in I/O speed doesn't really know what they're talking about, because never is compression going to give a 100% decrease in file size across the board, and it doesn't speed up the actual data transfer, just makes the amount of data being transfered take less time. There is a distinct but meaningful difference between those two concepts. Sometimes it will have an effect on performance, other times, it might not. There are too many variables to account for.

AngelicIceDiamond91d ago

The only closest examples that we're gonna see Is from upcoming cross gen games. The PS5 and X Series will load everything either instantly or in a few seconds with no pop ins with crisper visuals. Meanwhile the X1 and PS4 will be asleep at the wheel with loading. Best pick up PS5 or X Series this holiday IG you want the best cross gen games running as intended similar to beefy PC's. Can't go wrong with either console.

+ Show (1) more replyLast reply 91d ago
Futureshark92d ago

Is that a real photo of this guy?

timotim92d ago

Come on man, thats a Snapchat filter.

Marquinho92d ago

Legend says his beard is the source of his genius.

NeoGamer23291d ago

That is exactly how he looks. He has shown up on many interviews including the Digital Foundry XB hardware teardown exactly like that.

Father__Merrin92d ago

Ppl will always want an easy set top box for the TV to play games on

NeoGamer23291d ago

I agree completely. The console is all about the easiest (and mostly the best) way to play games on a TV in your home. I have multiple TVs and consoles on those TVs. A console makes it really easy for me and my family to play games and stream video content.

But, a console is not the only place gamers want to play games. I have PCs, tablets, phones, and laptops where I also like to play my games.