Jump to content

Which is the better quality silicone from these results? Did an enjoyable test.

Greetings,

I want to know how you can tell a silicon of a CPU is good or bad. From these test results.

 

My friend and I were curious about the whole silicon lottery and we both got the new 5600X CPUs, but his is from China and mine is from Malaysia. (little bit of googling shows that the silicon all come from USA). Anyways we wanted to find out

how much rng was involved and about silicon lottery. We tested both CPU on my PC, same parts and same bios settings.

 

First things to say are, both CPU had about the same temps recorded and max core clock speeds were basically the same. Max core clock speed read during these tests were 4651Mhz for both CPU and both CPU fluctuated between 4100 - 4250Mhz during the multicore run. Max temp read for both CPU was a 67C peak using a 240mm AIO.

 

The 5600X from China scored lower than the 5600X from Malaysia on Cinebench by ~5% points on average out of 5 runs for both multi and single core. However,  during those 5 runs, it pulled an average of 1.1V during the multicore run. Whereas, the one from Malaysia  pulled an average of 1.25V during the multicore run. 1.3V to achieve 4651Mhz for 5600X made in China and 1.34V to achieve 4651Mhz for 5600X made in Malaysia.

 

So I would like to know, which is considered the "better" silicon. The CPU (China) that can achieve about the same core clock speed with a lower voltage or the CPU (Malaysia) that manages to pull more voltage using the same settings and get slightly higher Cinebench scores.

Was interesting little test for me.

Link to comment
Share on other sites

Link to post
Share on other sites

3 minutes ago, QRCodeKiller said:

silicone silicone silicone silicone silicone

Silicon. Not silicone. Silicone is rubbery.

It's one of those things that irks me.

 

And the Malasyian one is better, as it is capable of higher performance.

elephants

Link to comment
Share on other sites

Link to post
Share on other sites

2 minutes ago, ragnarok0273 said:

Silicon. Not silicone. Silicone is rubbery.

It's one of those things that irks me.

 

And the Malasyian one is better, as it is capable of higher performance.

haha fixed.

 

interesting. for me I always thought that the CPU that achieves the same clock speed using a lower voltage would be considered having the better silicon. Any ideas why the one from China scores lower consistently despite getting the same clock speed at a lower voltage? They both performed in the same clock speed frequency range. Both fluctuated between 4124 and 4259Mhz during the multicore run. Just malaysia one pulled more voltage to achieve it. I'm just a super curious guy.

Link to comment
Share on other sites

Link to post
Share on other sites

Just because it comes from a different location doesn't mean one performs better, or has better silicon. they are likely using different silicon *batches* (which can come from the same place) it mostly just boils down to luck, and in this case, the malaysian one won.

AMD blackout rig

 

cpu: ryzen 5 3600 @4.4ghz @1.35v

gpu: rx5700xt 2200mhz

ram: vengeance lpx c15 3200mhz

mobo: gigabyte b550 auros pro 

psu: cooler master mwe 650w

case: masterbox mbx520

fans:Noctua industrial 3000rpm x6

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

7 minutes ago, Letgomyleghoe said:

Just because it comes from a different location doesn't mean one performs better, or has better silicon. they are likely using different silicon *batches* (which can come from the same place) it mostly just boils down to luck, and in this case, the malaysian one won.

Oh yeah, I've heard all the silicon comes from the same factory in the US and only packaging is done in China/Malaysia. My point wasn't where they are from, just wondering which one was better. I've always thought better silicon meant achieving same clock speed at lower voltage but this little test showed that while the Malaysian one had the exact same clock speed but had a higher voltage, literally the opposite of what i thought good silicon was, it performed better than the one that I thought had the better silicon.

Link to comment
Share on other sites

Link to post
Share on other sites

Malaysian sands is more pure, less polutions haha. Just kiddin.

Ryzen 5700g @ 4.4ghz all cores | Asrock B550M Steel Legend | 3060 | 2x 16gb Micron E 2666 @ 4200mhz cl16 | 500gb WD SN750 | 12 TB HDD | Deepcool Gammax 400 w/ 2 delta 4000rpm push pull | Antec Neo Eco Zen 500w

Link to comment
Share on other sites

Link to post
Share on other sites

4 hours ago, Letgomyleghoe said:

Just because it comes from a different location doesn't mean one performs better, or has better silicon. they are likely using different silicon *batches* (which can come from the same place) it mostly just boils down to luck, and in this case, the malaysian one won.

Sorry to bring this up again. Just so I understand what people mean by silicon lottery. The CPU that scores higher on Cinebench has the better silicon despite it seemingly be more inefficient than the other one? Also I'm curious, is the voltage that the CPU draw per core dependent on the silicon? ie. better silicon able to pull more voltage?

Link to comment
Share on other sites

Link to post
Share on other sites

14 hours ago, QRCodeKiller said:

The CPU that scores higher on Cinebench has the better silicon despite it seemingly be more inefficient than the other one? Also I'm curious, is the voltage that the CPU draw per core dependent on the silicon? ie. better silicon able to pull more voltage?

what better silicon means is it is able to achieve higher clocks, at the same voltage, or that it is able to achieve higher clocks at all. the voltage is however much you pump into it, you could put whatever voltage you wanted in any processor regardless of silicone quality

AMD blackout rig

 

cpu: ryzen 5 3600 @4.4ghz @1.35v

gpu: rx5700xt 2200mhz

ram: vengeance lpx c15 3200mhz

mobo: gigabyte b550 auros pro 

psu: cooler master mwe 650w

case: masterbox mbx520

fans:Noctua industrial 3000rpm x6

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

19 hours ago, ragnarok0273 said:

Silicon. Not silicone. Silicone is rubbery.

It's one of those things that irks me.

I made the same mistake earlier, guess I was typing too fast for my 30hz brain to keep up. I quickly rectified it. 😛 

 

Silicon quality is down to luck. Unless it's a special binned edition or something like that which is not the case here.

Link to comment
Share on other sites

Link to post
Share on other sites

3 minutes ago, Letgomyleghoe said:

what better silicon means is it is able to achieve higher clocks, at the same voltage, or that it is able to achieve higher clocks at all. the voltage is however much you pump into it, you could put whatever voltage you wanted in any processor regardless of silicone quality

Then could I ask why the one that draws higher voltage (Malaysia) is considered to have the better silicon? It gets the same clock as the other one but with a higher voltage. I'm not sure why it consistently scores higher than the other one that pulls less voltage. But both have same max clock speed, just voltage difference.

Link to comment
Share on other sites

Link to post
Share on other sites

5 hours ago, QRCodeKiller said:

Then could I ask why the one that draws higher voltage (Malaysia) is considered to have the better silicon? It gets the same clock as the other one but with a higher voltage. I'm not sure why it consistently scores higher than the other one that pulls less voltage. But both have same max clock speed, just voltage difference.

at stock settings, CPU clock speed and voltage are all over the place, so I'd assume it's hard to actually pinpoint the exact voltage that each CPU is running at. Given that both CPUs run at 67 degrees under load, at the same frequency, those voltage readings don't match up. A difference of 150mv will surely result in a significant load temperature difference. 

Why not overclock both chips to 4.7Ghz (as most people here manage to reach that clock speed) and see which one achieves it with a lower static voltage? That way you can for sure tell which chip has won the lottery, since both chips will be running with a fixed clock speed and voltage, thus eliminating variables.

 

Link to comment
Share on other sites

Link to post
Share on other sites

1 hour ago, Vals89 said:

at stock settings, CPU clock speed and voltage are all over the place, so I'd assume it's hard to actually pinpoint the exact voltage that each CPU is running at. Given that both CPUs run at 67 degrees under load, at the same frequency, those voltage readings don't match up. A difference of 150mv will surely result in a significant load temperature difference. 

Why not overclock both chips to 4.7Ghz (as most people here manage to reach that clock speed) and see which one achieves it with a lower static voltage? That way you can for sure tell which chip has won the lottery, since both chips will be running with a fixed clock speed and voltage, thus eliminating variables.

 

Actually I got bored and I did OC both to 4.7Ghz. So the CPU (China) that scored lower on Cinebench but achieved 4.65Ghz at lower voltage managed to OC to 4.7Ghz at 0.15V offset and managed to be stable during a 30min Cinebench run. The CPU (Malaysia) that scored higher on Cinebench but achieved 4.65Ghz at higher voltage managed to OC to 4.7Ghz at 0.175V and crashed instantly at 0.16875V offset.

 

Funny thing is that the CPU from China seems to have the better silicon as it can hit 4.7Ghz with lower voltage than the other CPU but when both CPU were stable at 4.7Ghz, the worse silicon CPU (Malaysia) still got higher scores in Cinebench than the "better" silicon CPU.

 

With the better silicon CPU, I also managed to get higher score with a lower offset voltage (0.15V vs 0.16875V). Very weird.

 

I tried 4.8Ghz with an offset of 0.35V but failed and gave up. So none of the 2 CPU could hit 4.8Ghz with a reasonable voltage.

 

 

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

×