Jump to content

power efficient CPU for gaming

Go to solution Solved by Fasauceome,

Third gen Ryzen processors are proposed to be far more efficient than current gen, you should consider waiting for the data on the new processors before you make a decision.

Just now, Islam Ghunym said:

you mean that hyperthreading will give more performance by raising required clock, so it won't change efficiency by itslef?

HyperThreading allows more instructions per clock by flooding the instruction queue with extra data, keeping it full, which is why it appears as additional cores in software. That extra load from the queued instructions increases power requirement, but not necessarily decreases "efficiency."

Link to comment
Share on other sites

Link to post
Share on other sites

1 minute ago, Islam Ghunym said:

you mean that hyperthreading will give more performance by raising required clock, so it won't change efficiency by itslef?

effect of hyperthreading is closer to adding cores rather than raising clock.

CPU: i7-2600K 4751MHz 1.44V (software) --> 1.47V at the back of the socket Motherboard: Asrock Z77 Extreme4 (BCLK: 103.3MHz) CPU Cooler: Noctua NH-D15 RAM: Adata XPG 2x8GB DDR3 (XMP: 2133MHz 10-11-11-30 CR2, custom: 2203MHz 10-11-10-26 CR1 tRFC:230 tREFI:14000) GPU: Asus GTX 1070 Dual (Super Jetstream vbios, +70(2025-2088MHz)/+400(8.8Gbps)) SSD: Samsung 840 Pro 256GB (main boot drive), Transcend SSD370 128GB PSU: Seasonic X-660 80+ Gold Case: Antec P110 Silent, 5 intakes 1 exhaust Monitor: AOC G2460PF 1080p 144Hz (150Hz max w/ DP, 121Hz max w/ HDMI) TN panel Keyboard: Logitech G610 Orion (Cherry MX Blue) with SteelSeries Apex M260 keycaps Mouse: BenQ Zowie FK1

 

Model: HP Omen 17 17-an110ca CPU: i7-8750H (0.125V core & cache, 50mV SA undervolt) GPU: GTX 1060 6GB Mobile (+80/+450, 1650MHz~1750MHz 0.78V~0.85V) RAM: 8+8GB DDR4-2400 18-17-17-39 2T Storage: HP EX920 1TB PCIe x4 M.2 SSD + Crucial MX500 1TB 2.5" SATA SSD, 128GB Toshiba PCIe x2 M.2 SSD (KBG30ZMV128G) gone cooking externally, 1TB Seagate 7200RPM 2.5" HDD (ST1000LM049-2GH172) left outside Monitor: 1080p 126Hz IPS G-sync

 

Desktop benching:

Cinebench R15 Single thread:168 Multi-thread: 833 

SuperPi (v1.5 from Techpowerup, PI value output) 16K: 0.100s 1M: 8.255s 32M: 7m 45.93s

Link to comment
Share on other sites

Link to post
Share on other sites

1 minute ago, RobFRaschke said:

There is some conjecture that T series are cherry picked just like K series are, but with little physical evidence that I've seen.

 

Higher baseclock requires more power, but is not necessarily less efficient. There are two means of measuring efficiency. Instructions per clock and performance per watt. Performance per watt is similar in function to performance per dollar. More in depth reviews like those I've seen an AnandTech assess Performance per watt.

can you pls gimme a link i can refer to, AMD also announced that their upcomming ryzen chips will have 15% better IPC than intel, will this mean they will be more efficient

Link to comment
Share on other sites

Link to post
Share on other sites

Just now, Jurrunio said:

effect of hyperthreading is closer to adding cores rather than raising clock.

so for example 6 cores 12 threads is as efficient as 8 cores 8 thread?

Link to comment
Share on other sites

Link to post
Share on other sites

2 minutes ago, Islam Ghunym said:

so for example 6 cores 12 threads is as efficient as 8 cores 8 thread?

in the case of 9700k and 8700k at matching frequency and voltage, yes. In other cases it might change

 

just keep in mind that regardless of core count and thread count, getting the same performance out from the same frequency and voltage leads to same level of power draw.

CPU: i7-2600K 4751MHz 1.44V (software) --> 1.47V at the back of the socket Motherboard: Asrock Z77 Extreme4 (BCLK: 103.3MHz) CPU Cooler: Noctua NH-D15 RAM: Adata XPG 2x8GB DDR3 (XMP: 2133MHz 10-11-11-30 CR2, custom: 2203MHz 10-11-10-26 CR1 tRFC:230 tREFI:14000) GPU: Asus GTX 1070 Dual (Super Jetstream vbios, +70(2025-2088MHz)/+400(8.8Gbps)) SSD: Samsung 840 Pro 256GB (main boot drive), Transcend SSD370 128GB PSU: Seasonic X-660 80+ Gold Case: Antec P110 Silent, 5 intakes 1 exhaust Monitor: AOC G2460PF 1080p 144Hz (150Hz max w/ DP, 121Hz max w/ HDMI) TN panel Keyboard: Logitech G610 Orion (Cherry MX Blue) with SteelSeries Apex M260 keycaps Mouse: BenQ Zowie FK1

 

Model: HP Omen 17 17-an110ca CPU: i7-8750H (0.125V core & cache, 50mV SA undervolt) GPU: GTX 1060 6GB Mobile (+80/+450, 1650MHz~1750MHz 0.78V~0.85V) RAM: 8+8GB DDR4-2400 18-17-17-39 2T Storage: HP EX920 1TB PCIe x4 M.2 SSD + Crucial MX500 1TB 2.5" SATA SSD, 128GB Toshiba PCIe x2 M.2 SSD (KBG30ZMV128G) gone cooking externally, 1TB Seagate 7200RPM 2.5" HDD (ST1000LM049-2GH172) left outside Monitor: 1080p 126Hz IPS G-sync

 

Desktop benching:

Cinebench R15 Single thread:168 Multi-thread: 833 

SuperPi (v1.5 from Techpowerup, PI value output) 16K: 0.100s 1M: 8.255s 32M: 7m 45.93s

Link to comment
Share on other sites

Link to post
Share on other sites

Just now, Islam Ghunym said:

can you pls gimme a link i can refer to, AMD also announced that their upcomming ryzen chips will have 15% better IPC than intel, will this mean they will be more efficient

A link to conjecture about cherry picking of T series and K series, not a specific one off hand.

 

I believe they said they +15% IPC over Zen+, not Intel. And they will be more efficient per core, but an IPC increase doesn't guarantee better efficiency on it's own. An IPC increase at the same wattage would be an efficiency increase. A die shrink and an IPC increase together will definitely increase efficiency. Efficiency requires Insructions/Performance per watt, which is a measure over time.

Link to comment
Share on other sites

Link to post
Share on other sites

4 minutes ago, RobFRaschke said:

HyperThreading allows more instructions per clock by flooding the instruction queue with extra data, keeping it full, which is why it appears as additional cores in software. That extra load from the queued instructions increases power requirement, but not necessarily decreases "efficiency."

so we don't know if it will decrease efficiency also...

Link to comment
Share on other sites

Link to post
Share on other sites

2 minutes ago, RobFRaschke said:

A link to conjecture about cherry picking of T series and K series, not a specific one off hand.

 

I believe they said they +15% IPC over Zen+, not Intel. And they will be more efficient per core, but an IPC increase doesn't guarantee better efficiency on it's own. An IPC increase at the same wattage would be an efficiency increase. A die shrink and an IPC increase together will definitely increase efficiency. Efficiency requires Insructions/Performance per watt, which is a measure over time.

why I cant find infos about CPUs effficiency anywhere, is that because customers usually never care about that?

Link to comment
Share on other sites

Link to post
Share on other sites

1 minute ago, Islam Ghunym said:

so we don't know if it will decrease efficiency also...

It's just not that simple. You're looking for a simple answer to a very complex question.

Link to comment
Share on other sites

Link to post
Share on other sites

2 minutes ago, RobFRaschke said:

It's just not that simple. You're looking for a simple answer to a very complex question.

but they can simply measure watts used and.... and try gaming benchmarks then take average and... XD it seems long a bit

Link to comment
Share on other sites

Link to post
Share on other sites

2 minutes ago, Islam Ghunym said:

why I cant find infos about CPUs effficiency anywhere, is that because customers usually never care about that?

Many of us do care about it, but it's difficult to measure effectively, and when it is measured, it's usually under an artifical load. Basically every AnandTech CPU review has a page dedicated to power useage with a chart that will look like this:

106299.png

 

The problem is, that this is under a specific, artificial full load, and not indicative of real world performance.

Link to comment
Share on other sites

Link to post
Share on other sites

6 minutes ago, RobFRaschke said:

Many of us do care about it, but it's difficult to measure effectively, and when it is measured, it's usually under an artifical load. Basically every AnandTech CPU review has a page dedicated to power useage with a chart that will look like this:

106299.png

 

The problem is, that this is under a specific, artificial full load, and not indicative of real world performance.

btw, is ryzen 5 2400G uses only 27W,that is mroe efficient than my intel shit one, I have gone for intel because i thought they will give me better power value, seems not! mm but boost clock is not considered here, seems really hard to get a good answer for this thx for help.

Link to comment
Share on other sites

Link to post
Share on other sites

in conclusion, all coffelake should have the same efficiency and by going for k processors I can get the best power value i want from any CPU of them, however zen 2 will be more efficient, cheaper and any zen2 cpu will be unlocked, so the final answer for me is to wait upcoming ryzen 3000 series, am I true here?

 

Link to comment
Share on other sites

Link to post
Share on other sites

1 hour ago, Islam Ghunym said:

dk but as rumored they are 7 nm power hungry chips. 

Uh, not really. 

 

MSI B450 Pro Gaming Pro Carbon AC | AMD Ryzen 2700x  | NZXT  Kraken X52  MSI GeForce RTX2070 Armour | Corsair Vengeance LPX 32GB (4*8) 3200MhZ | Samsung 970 evo M.2nvme 500GB Boot  / Samsung 860 evo 500GB SSD | Corsair RM550X (2018) | Fractal Design Meshify C white | Logitech G pro WirelessGigabyte Aurus AD27QD 

Link to comment
Share on other sites

Link to post
Share on other sites

1 minute ago, Stormseeker9 said:

Uh, not really. 

I was wrong probably, having bad english with hard time to understand things proberly, thx

Link to comment
Share on other sites

Link to post
Share on other sites

2 minutes ago, Islam Ghunym said:

I was wrong probably, having bad english with hard time to understand things proberly, thx

Quote

The AMD chip also consumed considerably less power—around 26 percent less, peaking at just over 133W versus the Core i9-9900K bumping up against 180W.

 

"Now you really see the power of 7nm technology and what being aggressive with technology does," Dr. Su noted.

 

 

 

MSI B450 Pro Gaming Pro Carbon AC | AMD Ryzen 2700x  | NZXT  Kraken X52  MSI GeForce RTX2070 Armour | Corsair Vengeance LPX 32GB (4*8) 3200MhZ | Samsung 970 evo M.2nvme 500GB Boot  / Samsung 860 evo 500GB SSD | Corsair RM550X (2018) | Fractal Design Meshify C white | Logitech G pro WirelessGigabyte Aurus AD27QD 

Link to comment
Share on other sites

Link to post
Share on other sites

1 hour ago, _Syn_ said:

I would also recommend the i5-9400F if you have a discrete GPU, shouldn't consume more than 60W,

from what I found there should be no difference because  IGP should turn off completely when it is not being used

Link to comment
Share on other sites

Link to post
Share on other sites

7 hours ago, Islam Ghunym said:

from what I found there should be no difference because  IGP should turn off completely when it is not being used

Yes that's true, it's just that the 9400F is more widely available and cheaper that's why i chose it, it's 155$ on Amazon, but be aware that you need to update your BIOS before you install it if you wanna go that route

Quote or Tag people so they know that you've replied.

Link to comment
Share on other sites

Link to post
Share on other sites

2 minutes ago, _Syn_ said:

Yes that's true, it's just that the 9400F is more widely available and cheaper that's why i chose it, it's 155$ on Amazon, but be aware that you need to update your BIOS before you install it if you wanna go that route

But the 9400F is not a worthy upgrade, so I would really suggest waiting for Ryzen 3000 and see what it offers

Quote or Tag people so they know that you've replied.

Link to comment
Share on other sites

Link to post
Share on other sites

Here's wild suggestion – why not get Ryzen 5 APU and ditch dedicated graphics card? 
I am not sure how good 1060 is when it is undervolted to reach 60W, going with APU would allow you lower power consumption

Ex-EX build: Liquidfy C+... R.I.P.

Ex-build:

Meshify C – sold

Ryzen 5 1600x @4.0 GHz/1.4V – sold

Gigabyte X370 Aorus Gaming K7 – sold

Corsair Vengeance LPX 2x8 GB @3200 Mhz – sold

Alpenfoehn Brocken 3 Black Edition – it's somewhere

Sapphire Vega 56 Pulse – ded

Intel SSD 660p 1TB – sold

be Quiet! Straight Power 11 750w – sold

Link to comment
Share on other sites

Link to post
Share on other sites

13 hours ago, Stormseeker9 said:

 

 

...and that is also rumored to be on an X570 Plattform, wich might consume a bit more.


Some people assume that the CPU itself might actually be around  half the Intel. Wich is possible s the other components are the same and the MoBo might not be the most efficient one...

"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

6 hours ago, _Syn_ said:

But the 9400F is not a worthy upgrade, so I would really suggest waiting for Ryzen 3000 and see what it offers

Yeah, totally agree. And also Picasso, if the APU is an option.

"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

On 5/18/2019 at 4:15 PM, Quadriplegic said:

Here's wild suggestion – why not get Ryzen 5 APU and ditch dedicated graphics card? 
I am not sure how good 1060 is when it is undervolted to reach 60W, going with APU would allow you lower power consumption

well, my current sample is making impressive results at 60W only, it can do 3/4 stock performance at furmark and usualy gives 70-90% of stock performance on games running at 1400 Mhz booting to 1635 MHz a I monitored and with it's GDDR5.... the APU can in no way get a similar performance, however the NAVI 20 will be integrated in a 95W TDP Ryzen 5 3600G which may give an idea about being power hungry since even the Ryzen 5 3600X is expected to be with lower thermal design and of course I will have to overclock the IGP to get an acceptable performance which will make it less efficient, however all these are just rumors, but I see it will be nearly true.

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

×