Jump to content

Intel Roadmap Leaks: 9th Gen Sept, New, 28 core A-Series

2 minutes ago, porina said:

AMD intentionally didn't put in as much FPU when designing Zen.

That is only true when talking about 256bit AVX code, with "normal FPU" Code the difference isn't that much.

 

But most of the power consumption measurements are bullshit because of the power saving mechanism things...

 

So you have to record the Power Consumption over time and compare it with the work done anyway. There is no other way these days, wich makes it just more complicated....

"Hell is full of good meanings, but Heaven is full of good works"

Link to comment
Share on other sites

Link to post
Share on other sites

Wait. I'm confused...

 

Where's the leak in the roadmap?

Please quote me or @ me in your response so I get a notification.                                                                       I really really really really like small text.

Occupation: Gamer | Hours: 24/7 Full-Time | Hourly Pay: 0 | Benefits: Chick Magnet

 

 

 

FINISHED BUILD - VIEW PROFILE

 

 

 

If you say I'm not always right, but I am, I will say I am right. 

rekt

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

1 minute ago, Stefan Payne said:

That is only true when talking about 256bit AVX code, with "normal FPU" Code the difference isn't that much.

I've seen similar differences in x87 code too. 

 

1 minute ago, Stefan Payne said:

But most of the power consumption measurements are bullshit because of the power saving mechanism things...

I'm more interested in heavy loads, ones that are pretty continuous, so you could avoid it that way.

 

1 minute ago, Stefan Payne said:

So you have to record the Power Consumption over time and compare it with the work done anyway. There is no other way these days, wich makes it just more complicated....

I never said it would be easy, which is in part why it isn't so likely for people to do this type of testing. I know some places monitor the power delivered to the CPU connector on mobo, which is a start. If I were to do it, I'd be limited to self reported power usage, and I have no idea how accurate either AMD or Intel are in that area.

Main system: i9-7980XE, Asus X299 TUF mark 2, Noctua D15, Corsair Vengeance Pro 3200 3x 16GB 2R, RTX 3070, NZXT E850, GameMax Abyss, Samsung 980 Pro 2TB, Acer Predator XB241YU 24" 1440p 144Hz G-Sync + HP LP2475w 24" 1200p 60Hz wide gamut
Gaming laptop: Lenovo Legion 5, 5800H, RTX 3070, Kingston DDR4 3200C22 2x16GB 2Rx8, Kingston Fury Renegade 1TB + Crucial P1 1TB SSD, 165 Hz IPS 1080p G-Sync Compatible

Link to comment
Share on other sites

Link to post
Share on other sites

28 minutes ago, porina said:

I never said it would be easy, which is in part why it isn't so likely for people to do this type of testing. I know some places monitor the power delivered to the CPU connector on mobo, which is a start. If I were to do it, I'd be limited to self reported power usage, and I have no idea how accurate either AMD or Intel are in that area.

This is still not perfect as it is affected by VRM efficiency, technically the best way to get raw CPU power usage is to monitor voltage and current going to the CPU over x period of time, but that's not always possible.

Just some bapo nerd from 'Straya

 

PCs:

Main: i7 7700K (5GHz 1.4V) | ASUS GTX 1080 TURBO | 4x8GB Corsair Vengeance 3000MHz (3200MHz CL14 1.365V) | ASUS PRIME Z270-AR | Thermaltake SMART 750P | Coolermaster Seidon 240P | Acer Predator X34 (34" 1440p144Hz GSync IPS)

 

Secondary: i5 3570K | Intel HD4000 (RIP Sapphire HD 6850) | 2x2GB + 1x4GB Kingston 1600MHz | ASUS P8Z68-V LX | Corsair CX650 | Coolermaster Hyper D92 | Sony Bravia VPL-VW80 (108" 1080p60Hz projector)

 

Laptop: i7 7700HQ | GTX 1060 6GB MXM | 2x16GB SODIMM | OEM Acer Motherboard | 17.3" Screen (1080p60Hz IPS)

 

iMac: Core 2 Duo T7400 | ATI Radeon X1600 | 2x1GB 667MHz DDR2 | 20" Screen

Link to comment
Share on other sites

Link to post
Share on other sites

3 hours ago, awesomegamer919 said:

This is still not perfect as it is affected by VRM efficiency, technically the best way to get raw CPU power usage is to monitor voltage and current going to the CPU over x period of time, but that's not always possible.

its not possible because although voltages are easy to get currents is much harder, it could be possible to do by the mobo manufacturers by adding a loop around the power plane though even that might not be doable due to how crowded that area of the board is 

the best way is to figure out the vrm efficiency by looking at the vrm datasheets and then reducing that from the power going into the the motherboard (much easier to measure)

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×