• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Do you think consoles should be more costly and powerful or like the wii?

Page 3 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
i see two possibilities. first, the next xbox/ps4 will be at least twice as powerful as this gen. this is the more 'traditional' method. bigger, better, faster, more expensive etc.
the second, and less likely, is doing some kind of super cheap box as a delivery service. kind of like onlive. but something tells me they'll go more traditional since that would be too radical approach.
 
Market segmentation is a very bad thing. Just compare what the consoles can to *today* with 2005/2006 tech, and what kind of graphics/framerates you could get out of 2005/2006 PCs. There's a fair amount of overhead involved with PC games, so you can't expect true feature parity, but console devs have the advantage of optimizing for one platform, so you see boundary-pushing titles like Skyrim working on consoles, whereas the PC version will require significantly better hardware to match the console experience.

Exactly. People need to understand this.
 
I'm going to go ahead and be a spec-whore by listing the specs of my dream console so critique the following (I know that it would be expensive if it were possible, and that having purely programmable rendering (other than the textures) would be a challenge for the game devs, but it would certainly be rewarding for the best programmers)...

CPU: IB/SB derivative:
[16 nm]
4 cores with hyperthreading (so 8 threads) @ 3.2 GHz
intergrated VRM
integrated PCH (SATA controller with 2 eSATA (1 passive, 1 active) 300MBps ports, 1 GbLAN contoller for ethernet port, 1 S/PDIF optical transmitter[1], USB 2.0 controller [2])
PCI-E 3.0 x16 link to GPU
PCI-E 3.0 x1 link to expansion slot 1
PCI-E 3.0 x16 link to expansion slot 2
8 MB L3 cache with not more than 1/2 the latency of a 2600K's L3 cache.
L2 and L1 cache would also have not more than 1/2 the latency of a 2600K's cache
[1]2 channels, 16b: 22.05/33.075/44.1/48/88.2/96/176.4/192 KHz
2 channels, 24b: 48/96/192 KHz
[2] 4 regular USB 2.0 ports (default 500 Hz/port; 2KHz/port tolerance)
4 proprietary USB 2.0-based controller ports (default 500 Hz/port; 2KHz/port tolerance)

4 GB (2GB/channel) DDR3 @ 1.333 GHz with 7-7-7-21-1T timings

GPU:
[12 nm]
8 Pentium MMX (derivative) cores with 8 custom 1024 bit wide AVX (2.0 +) units/core @ 2 GHz
[1 MB cache/core]
single, double, and double extended (80 bit) FX and FP precision formats.
TSX, FMA 16
128 (16/core)/512 (64/core) TA/TF units @ 600 MHz
[1 MB texture cache]
lossless map compression, bindless textures, same HQ filtering quality as Fermi; no lossy map formats
RGBA10FX/10FP/16FP/20FP (1/2 speed)/32FP (1/2 speed) lossless map formats
integrated TMDS and latest version of video decoder; same as Kepler TMDS (including, but not limited to 1:1 pixel mapping/FX AR scaling/same AR scaling, user selectable color space and user adj brightness, gamma, contrast, and vibrance) other than support for lightpeak and no support for audio.
0/1/2/3/4/5/6/7/8 max frames to render ahead
frame rate limit (custom)
24-240 Hz custom refresh rate.
2 video multiouts (supports dual displays and teaming; uses a proprietary dongle that fits into to the video multiout port securely for each output type):
DL-DVI
Displayport (latest version, except without audio support)
Lightpeak
[18 MB L2 cache]
576 bit 4 GHz GDDR5 IMC
(9/2) GB GDDR5

I'd pay to have the consumer models manufactured by Asus with all japan made caps and fingerless, padded Q shields for where the I/O ports were; 2 oz of copper would be used for the PCB; standoffs would be installed in the console's casing so that nothing short circuits.

The power supply would have .125% max ripple p-p; - to + 3% max deviation on the 12V rail with 115/230 VAC input using 14 AWgauge copper power cord and 14 gauge copper house wire. It would have an 80 mm ball bearing exhaust fan with PWM control for 16-48 cfm. The single 12V rail would be 76 A continuous. It would have all made in japan caps, all copper wiring/interconnects, and heavy duty DC->DC conversion. Finally, it would have SCP, and protection against too low and too high of input voltage.

I'd have a single twister bearing fan (180mm, 1.25-2.5 krpm) at the top for exhaust; running at stock voltage - to + 3% maximum

All copper heatpipes and bases, and all aluminum fins for cooling the CPU and GPU dies.

I'd use AC MX-2 in all places that TIM is applicable.
 
Last edited:
i see two possibilities. first, the next xbox/ps4 will be at least twice as powerful as this gen. this is the more 'traditional' method. bigger, better, faster, more expensive etc.
the second, and less likely, is doing some kind of super cheap box as a delivery service. kind of like onlive. but something tells me they'll go more traditional since that would be too radical approach.

Streaming will be the eventual future. But with the cost required to get the speed and data caps necessary, a lot of people won't buy into a streaming only console. Netflix HD already chews through 2GB per hour. It won't happen this next generation but likely the one after. Though you may see Sony and Microsoft start experimenting with it. Everyone in the industry, the big boys anyway, want a system that changes games from commodity to service. Keeps you paying to keep playing.
 
Back
Top