How long will it take for PC to catch up to Xbox One X?

I think PCs already have. And this is from someone who's pretty excited about the X1X.

oh, I realize that.. for the most part you've been quoting comments from developers or Digital Foundry.. cows don't seem to understand what "context" means these days though so, in fairness, I had to mention your name..

Pc´s as already caught up.

The Xbox one X is nothing new , and while it´s a powerful console.

If you have a decent pc , there is no reason to get a Xbox one X.

Anything that comes on the xbox will come to the pc.

No one's doubting that PC can be more powerful that the X1X

@Jacanuk: So you have a $500 PC with comparable performance?

:P

NO i was joking but you are so into defending the xbox one X that you simply didn't get it..hahahahahaa

You really are a machine or simply lack the most basic knowledge about conversational English.

You actually took my 600 years joke as true and you refute it....Hahahahahaaha

Hahahahaa transistors wars...lol

Last i remember there was a console with 5 billion transistor that got spanked by another console with much less transistors..lol

The RX580 is just a GPU the SOC for scorpio is 7 billion transistors they are not comparable,the same crap was say about the xbox one already with its 5 billion transistor crap...lol

There's no way there's a 1070 in the X1X that thing cost roughly 400 or more depending on where you shop. In order to put something like that in and make a profit Microsoft would have to charge roughly 800

He's incapable of understanding that some things should not be taken literally. I understand now why he has so much trouble with human communication.

You can't use the full extent of XB1X's hardware without XBL.

You will start saving money on a PC after 3-4 years (and every gen after that), due to being able to upgrade vs buying XB2X + XBL.

But yes, as a 100% brand new PC gamer it will be more costly at first.

Entirely new superior products cost more. Double the fps alone is worth it.

what part of its hardware is inaccessible?

@xboxiphoneps3: No, no, no, and no.

The market will readjust for miners. Either when supply meets demand, or when the inevitable amount of miner abused cards flood the second hand market. Or the mining industry becomes inaccessible to amateurs once again (as bitCoin did).

Even when AMD wasn't competitive the next generation x60 always out performed the last generation x70 cards, as far back as Kepler.

Volta might be delayed till late next year, but it will outperform the X1X on the cheap. As will the next gen i3's and Pentiums.

Maybe 1-3 years? I wonder how powerful some AMD ravenridge or its successors would be just for gaming performance.

I do.

In fact i have an i7, i5, and a 980ti in the i5 and in total the two systems and the card were under 500 (systems were not complete, but mobo, processor, some ram, etc. the rest i had lying around). The i5, while i dont really give a shit about 4k gaming, i know it can do it. The average would probably be 28-34fps, but it can. If i put it in the i7 i'd probably get 3-5 extra frames.

The thing is. Computers depreciate in value.

So even though I paid X amount for my computer 2 years ago and it would easily whip the shit out of an X1X, it does not mean it is worth the same value today as when I bought it.

Just proves this whole comparison is complete garbage.

So when do you think you could build a new system without using "parts laying around"?

Also, the 980 would cry uncle very suddenly in 4K, 4 GB of vram isn't enough.

Incoming RX-Vega 36/40/44(??) or RX-590 (Vega 11) to take on GTX 1070. Vega 11 is something faster than RX-580.

Vega 56 ~= GTX 1070 Ti

Vega 64 ~= GTX 1080

There's little need for NV to move towards Volta which is effectively Pascal with FP16 RPM.

GP102 already has double rate INT16 RPM.

Your post is a spam which has nothing to do with system wars but engages in poster wars.

Another knee-jerk reaction from a cow who engages in poster wars instead of system wars.

Stock 980 Ti has 6GB VRAM with 336 GB/s. Kod has GTX 980 TI not the lesser GTX 980. Super OC 980 Ti can beat super OC 1070 at 4K.

Wrong. You didn't read the article.

A Custom AMD GPU

One of the biggest draws of the Scorpio is its graphical capabilities. The GPU was custom designed with AMD, featuring 40 compute units clocked at 1172MHz. With a reported 7 billion transistors, the GPU is capable of 6 Teraflops of processing. For comparison, the last generation of Xbox consoles had 1.32 Teraflops of power.

X1X's GPU has more CU, more hardware features and caches when compared to RX-580. (5.7 billion transistors / Polaris 36 CU) x retail X1X's 40 CU = 6.3 billion transistors.

X1X's GPU has more hardware features and caches when compared to R9-390X (6.2 billion transistors).

From http://www.anandtech.com/show/11740/hot-chips-microsoft-xbox-one-x-scorpio-engine-live-blog-930am-pt-430pm-utc#post0821123606

12:36PM EDT - 8x 256KB render caches

12:37PM EDT - 2MB L2 cache with bypass and index buffer access

12:38PM EDT - out of order rasterization, 1MB parameter cache, delta color compression, depth compression, compressed texture access

X1X's GPU's Render Back Ends (RBE) has 256KB cache each and there's 8 of them, hence 2 MB render cache.

X1X's GPU has 2 MB L2 cache, 1 MB parameter cache and 2MB render cache. That's 5 MB of cache.

https://www.amd.com/Documents/GCN_Architecture_whitepaper.pdf

The old GCN's render back end (RBE) cache. Page 13 of 18.

Once the pixels fragments in a tile have been shaded, they flow to the Render Back-Ends (RBEs). The RBEs apply depth, stencil and alpha tests to determine whether pixel fragments are visible in the final frame. The visible pixels fragments are then sampled for coverage and color to construct the final output pixels. The RBEs in GCN can access up to 8 color samples (i.e. 8x MSAA) from the 16KB color caches and 16 coverage samples (i.e. for up to 16x EQAA) from the 4KB depth caches per pixel. The color samples are blended using weights determined by the coverage samples to generate a final anti-aliased pixel color. The results are written out to the frame buffer, through the memory controllers

GCN version 1.0's RBE cache size is just 20 KB. 8x RBE = 160 KB render cache (for 7970)

AMD R9-290X/R9-390X's aging RBE/ROPS comparison. https://www.slideshare.net/DevCentralAMD/gs4106-the-amd-gcn-architecture-a-crash-course-by-layla-mah

Render Back Ends includes it's own math operations i.e. "Logic Operations" and Blending.

16 RBE with each RBE contains 4 ROPS. Each RBE has 24 KB cache.

24 bytes x 16 = 384 KB.

X1X's RBE/ROPS has 2048 KB or 2 MB render cache.

X1X's RBE has 256 KB. 8x RBE = 2048 KB (or 2 MB) render cache. X1X has hold more rendering data on the chip when compared to Radeon HD 7970.

http://www.eurogamer.net/articles/digitalfoundry-2017-project-scorpio-tech-revealed

We quadrupled the GPU L2 cache size, again for targeting the 4K performance."

X1X GPU's 2MB L2 cache can be used for rendering in addition to X1X's 2 MB render cache.

RX-480's RBE (Render Back Ends)

It seems RX-480's RBEs doesn't have any improvements.

The following block diagram shows RX-480/RX-580's RBE(Pixel Engine) bottleneck!

X1X's RBE/Pixel Engine has 2MB render cache and it's located next 2MB L2 cache!

When X1X's L2 cache (for TMU) and render cache (for ROPS) is combined, the total cache size is 4MB which is similar VEGA's shared 4MB L2 cache for RBE/ROPS and TMUs.

X1X GPU also has new workload distribution and "conservation occlusion query".

https://msdn.microsoft.com/en-us/library/windows/desktop/dn914594(v=vs.85).aspx

Conservative Rasterization

Overview

Conservative rasterization means that all pixels that are at least partially covered by a rendered primitive are rasterized, which means that the pixel shader is invoked. Normal behavior is sampling, which is not used if conservative rasterization is enabled.

Conservative rasterization is useful in a number of situations, including for certainty in collision detection, occlusion culling, and visibility detection.

X1X's version got the core functionality from Conservative Rasterization.

Chill out, ron

@ronvalencia: "Another knee-jerk reaction from a cow who engages in poster wars instead of system wars."

If you can't communicate like a human being it's difficult to argue about subjects. Your robotic-spamming style creates the "personality wars".

Never. Im always buying computer shit and have tons of it laying around. But the cpu, gpu and mobo are the most expensive parts. Even if i didnt have those other parts, in total it might have cost an extra 100-150 bucks for both systems.

980ti's have no problems running 4k at 30fps. Yes, if you want to go above that you will have problems, but whatever. 4k is a silly benchmark for consoles or PCs to be honest.

Oh, and Ron already addressed the 6gb thing. But i'd be willing to bet 4gb is enough for 4k and it would probably be on par with consoles.

EDIT - https://www.extremetech.com/gaming/213069-is-4gb-of-vram-enough-amds-fury-x-faces-off-with-nvidias-gtx-980-ti-titan-x

but I spent 94$ on mine. https://www.newegg.com/Product/Product.aspx?Item=N82E16813128988

@kod: Right, but you originally said 980, not 980 ti. You can only run 4K on 4GB of ram with really crappy texture settings.

The X1X will match PC ultra settings, and 30 fps is ok with that level of image quality. Pretty sure the human eye can't see more than that anyhow, which is why videos are recorded at 24 fps.

Nah, looks like i did say 980ti.

The human eye caps out at 120fps (its actually like 105 or 110 but we just go with 120 since its an even fps measurement) and IMO, fps is far more important to the gaming experience than doing 4k over 1080. The difference between the two for most gaming set ups is negligible. Where-as increased fps affects graphics, controls, even sound (sync). Im not quite sure what videos you are referencing as you can choose what fps you record at with most digital cameras.... i suspect you're referencing streaming, probably a base youtube setting? I dont know, but camera's can record up to like 40,000 fps if you get a special one, a normal one will do at least 60.

There is no reason for any gaming system to be attempting anything less than 60fps right now. I can almost guarantee you that if you played a game at 4k 30fps and then played the same game at 1080 but 60fps, you'd never go back to the 4k. Because of the split second responses, i feel this is especially true when it comes to fighting games. But once you realize how important at least 60fps is to controls and smoothness of a game, going with something less (even if 4k... hell 8k.... shit 100,000k) is like trying to go back to 56k modems from cable modems.

@ronvalencia said:

@tormentos said:

Hahahahaa transistors wars...lol

Last i remember there was a console with 5 billion transistor that got spanked by another console with much less transistors..lol

The RX580 is just a GPU the SOC for scorpio is 7 billion transistors they are not comparable,the same crap was say about the xbox one already with its 5 billion transistor crap...lol

Wrong. You didn't read the article.

X1X's GPU has more CU, more hardware features and caches when compared to RX-580. (5.7 billion transistors / Polaris 36 CU) x retail X1X's 40 CU = 6.3 billion transistors.

X1X's GPU has more hardware features and caches when compared to R9-390X (6.2 billion transistors).

You are not READING...

The xbox one X 7 billion transistors is from its complete SOC not just for the GPU,so making comparison is stupid,is clear the GPU has more transistors than the RX480 and 580 but the comparison are lame The RX580 is 6+TF with 36CU because of its speed regardless of having less transistors.

The Scorpio Engine is the new system on chip (SoC) that lies at the heart of the machine- it is a 16nm FinFET chip produced by TSMC, with seven billion transistors containing a custom CPU cluster and 40 customized Radeon compute units- among other things.

http://gamingbolt.com/xbox-scorpio-internal-motherboard-architecture-revealed-16nm-finfet-chip-7-billion-transistors-vapour-chamber-cooling-and-more#3VpeqMb74RcaARDQ.99

And yes i am right.

@ronvalencia:

That's a lot of extra cache. Looks like they're using tile based rendering and have direct pixel engine to L2 access. Nothing else would really explain the need for that amount.

So a game like DOOM should have decent performance on 6TF RX 480 in comparison, while traditional Nvidia stronghold/gimpworks titles should be much faster on X1X. This is explains why some titles scale much better on X1X than PS4 Pro, up to 100% better, as opposed to the 43% you'd expect.

---

X1X is shaping up to the be the greatest console hardware ever assembled.

How long will it take for PC to match the Xbox One X in power, features, performance at the same price point? Not today, not today!

The issue is XB1X already can't reach PC. 30 fps Destiny 2 and XBL which is far worse than PC online.

Two to three(2 years) more GPU generations... You will get a RX760/GTX3050Ti with similar performance to a RX 580(590ish) for $200 but by then the X1X will probably be sold for $399.

The X1X in terms of price/performance is probably the best console in a long time.

That being said PC will always be more powerful, its no win battle. You guys get consoles built for price/performance where the manufactures make deals and lots of cost cutting is done its to meet the desired performance... PC can never compete directly since components are sold individually.

If you want price/perfomance and dont care about framerate get a X1X... If you care about framerates get a PC with a high end GPU, its simple as.

Everyone knows PC can surpass xbox

Xhawk never stated price, and he mentioned features, in which PC is better even if it runs worse. Also online is a huge feature - meaning XBL charges are part of his post meaning PC is cheaper.

"performance at the same price point?"

Yeah like the PS4 didn't have gaps over 100% with just 40% more power..Boy how fast they forget resolution gate..lol

IN fact doom on PS4 is almost always 1080p and almost always 60FPS,on xbox one drop to 40's frame wise and resolution drops to lower than 828p which combined does make for a bigger than 100% gap.

Meh... was going to adjust this but whatever.

So to add to Zaryia, a lot of people never consider these extra costs when it comes to consoles. Now, ive been a console and PC gamer most of my life, the only time i stopped caring about consoles was with this 8th generation when they decided to do nothing aside from being low grade gaming PCs. And i have bought plenty of systems at launch and the things people tend to leave out when mentioning price are all the other things youll need. So with this XBX1 or whatever it is, its not just 499 and you're good. Its 499+tax, + paying the unnecessary pimp that is xbox live + another controller (or some other peripheral that youll regularly use) + 2 or 3 games = You're easily looking at 750. So at this price point, i would be able to build a PC that consoles most likely wouldnt even be able to touch.

I really hate to say it given that ive been a gamer for over 30 years, but these new consoles are nothing but a practice in bad consumerism.

@tormentos:

Who the **** is "they" and why are you bringing the PS4 into this?

---

The difference here is that X1X is not only demonstrating a larger resolution gap in relation to the Pro, but it's also increasing the settings at the same time.

Digital Foundry did benchmarks on a batch of faceoff games and they found the relative gap between the PS4 and XO was lower than the theoretical gap. It won't be like this with PS4 Pro vs X1X because MS has addressed memory bandwidth limitations while Sony has not.

The compute gap(i.e. TFLOPS) will mimic theoretical percentage on games with compute path optimization like DOOM(2016), but in titles that favor Nvidia(i.e. most titles), the X1X will likely have some form of tile based rendering paired with a ton of cache and direct pixel to L2 cache connection, with 6-7GB VRAM at 326GB/s on a 384-bit bus. Pro is locked behind old GCN memory bandwidth limitations and memory controller bottleneck. The gap will be larger than the theoretical amount, as opposed to PS4 vs XO. This is more Nvidia vs AMD general performance relative to TFLOPS than it is XO to PS4.

Nope. Bungie made a deal with Sony so that is the reason for 30fps on X1X. Second features like dolby atmos and 4k 60fps hdr caputure on the Xbox One X plus 4k UHD blu-ray drive.

Nope. 60 fps > 30 fps

http://gamingbolt.com/xbox-one-xps4-pro-graphical-differences-will-be-visible-in-big-aaa-games-betadwarf

Yeah there is going to be a big difference on AAA games.

@xhawk27:

TW3 is a game that will be a good benchmark since even though it's a gameworks title, it's got pretty even performance at the GTX 1060/RX 480 price point. I've spent a decent amount of time benchmarking it and I expect X1X to beat my GTX 1060 and RX 480 at max OC by around 18%.

Vega 11 arrives when X1X drops in...

Vega 11 to replace RX-580 (36 CU) and NV placing GTX 1070 Ti (18 SM) into 1070 (15 SM)'s slot hence moving 1070 downwards. AMD could keep RX-580 for the miners (they don't need RBE updates) while Vega 11 for the gamers.

PC world is moving again.

DOOM used heavy Compute Shader based rendering hence enables RX-480's TFLOPS to work better with Polaris 10/20's 2MB L2 cache. Significant amount of NV Gimpworks uses Pixel Shader path.

I'd settle for Xbox One S over X if you wanna get yourself a cheap ass 4K/HDR movie set.

Nope, not even a single generation.

Vega 11 has RBE updates like X1X's RBE updates. Think of RX-580 with Vega 64 era clock speed e.g. 1.6 Ghz (7.3 TFLOPS with 36 CU) and RBE with mulit-MB of cache (copies GP104's L2 cache layout). RX-580 already has comparable triangle rate as GTX 1070's 5G and it's just missing RBE updates and DX12 Feature 12_1(rasterization feature updates, Pascal feature level).

https://www.pcgamesn.com/amd-vega-11-production

"AMD’s Vega isn't finished yet, Vega 11 goes into production to replace Polaris".

AMD has two Vega generations and these are 14 nm FinFET and 14 nm FinFET+ updates before 7 nm FinFET Navi arrives.

Vega 11 is needed for Raven Ridge SoC's Vega based IGP partnering e.g. a gaming laptop with Vega 11 (~36 CU) dGPU + SoC's Vega based IGP (11 CU) hybrid setup. Laptop market is about half of X86 market..

The problem is the miners and hopefully AMD keeps RX-580s for the miners while gamers has Vega 11.