Post your Cinebench R20+15+R11.5+2003 Scores **Don't Read The OP PLZ**

[Jumper118]

not my fault nobody is posting scores any more  seems we got the most subs when haswell was around which is sad cos they were pretty boring

[ViBEE]

5820k@4624Mhz/3125Mhz, Kingston HyperX Savage 2x8GB@3333mhz 16-17-17-36-430-2T

 

 

[done12many2]

On 12/9/2017 at 3:50 AM, Jumper118 said:

not my fault nobody is posting scores any more  seems we got the most subs when haswell was around which is sad cos they were pretty boring

 

Did you get your TR4 water block in?

[Princess Luna]

Going to leave here as this is the last weekend I am in possession of my old already sold build featuring:

 

i7 6700

Asus H110m-C

8gb Corsair Value Select 2133mhz Sold with the system but already was using my brand new Trident Z 3200mhz issue is the motherboard that limits it to 2133mhz 15 15 15 36 and I believe this was the major slowing down issue with this ultra budget i7 experience.

TITAN X Maxwell Sold and replaced by EVGA 1080 Ti FE

 

CPU cooled by Stock Cooling, yes I know kill me now

Spoiler

Not bad performance considering it was cheaper than going i5 6600k back in the day by every mean.

 

Posting this here to be able to compare with what results I get from the i7 8700 with actually overclocked memory next week, also with more adequate cooling provided by the Cooler Master 212X

[Masada02]

On 12/9/2017 at 2:50 AM, Jumper118 said:

not my fault nobody is posting scores any more  seems we got the most subs when haswell was around which is sad cos they were pretty boring

So what you're saying is I should bench my SFF 8700k build. K...

 

[Jumper118]

On 12/5/2017 at 10:50 PM, Jumper118 said:

 

XSPC block better than air  single channell is still crap as ever though. I 1000% blame the memory for this score not being over 3500cb

 

On 12/10/2017 at 8:50 PM, done12many2 said:

 

Did you get your TR4 water block in?

 

[done12many2]

11 minutes ago, Jumper118 said:

XSPC block better than air  

 

It's always great when you state the obvious.  

 

11 minutes ago, Jumper118 said:

single channell is still crap as ever though. I 1000% blame the memory for this score not being over 3500cb

 

Yeah, it's 100% the memory holding it back.  3.9 to 3.925 GHz with memory in quad at 3600 MHz running tighter primaries and Trfc will go into the low 3400cb range.  Latency on the RAM is what's really killing Ryzen.  BADLY.  Bandwidth is great, but It takes a lot of work to get latency down into the 70ns and if lucky the 60ns range, which still sucks!  Basically, on the memory side, it's a shit show.

 

At least you got a chip that can run CB at 4.1 GHz.  I'm seeing some poor bastards that can't run hardly anything above 4 GHz.  I suspect noob factor on a lot of that.  

 

I've also come to the realization that Vcore requirements on Ryzen increases quite a bit compared to Intel as you make memory faster.  That boost from the memory puts a lot more data across the package and you'll need to up the voltage to keep it running.  The reward from the extra memory performance is great so it's worth it. 

[Herman Mcpootis]

@Jumper118

[Dabombinable]

I've got no problems running CB R11.5 on my Pentium M laptop with its Intel 915. This one with the newer 965/GMA X1300 however just causes it to crash. So here are the CB 2003 results for my Asus F3L, and god do I miss my Tecra M5-it was so much better.

[Pannalicour]

AMD Ryzen 1700 @ 3.940 GHZ (STOCKCOOLER!)  Memory: 2*4gb@ 3200MHZ DDR4

 

Cinebench r15 score 1754 points!

For a 290 dollar processor and a 80 dollar motherboard. Total 370 dollar.

 

I don't think anything can beat this price/performance at this moment in this (cpu/motherboard) market. What do you guys think?

 

 

[Lemon Wolf]

Latest result with my 1950X.

 

 

On 30.11.2017 at 11:18 PM, done12many2 said:

Small update to last submission.

 

TR 1950x with CPU @ 4.1 GHz / Mem @ 3600 MHz c15

 

Multi - 3631cb

Single - still sucks!!

 

To be honest the single threaded score doesnt suck at all.

Most Intel CPUs boost to quite high clock speeds when running single threaded applications.

When you test the single threaded performance at a fixed speed on both platforms the difference is minimal.

Btw how did you manage the 3600 on your ram? It doesnt matter what i try i cant get higher than 3500. Anything much higher than that results in boot failure.

[done12many2]

1 minute ago, Lemon Wolf said:

To be honest the single threaded score doesnt suck at all.

Most Intel CPUs boost to quite high clock speeds when running single threaded applications.

When you test the single threaded performance at a fixed speed on both platforms the difference is minimal.

 

Thanks for the break down on how it works, but I can run "fixed speed" on my Intel 7900x and score 220 + in single threaded runs (~30% higher at fixed clocks).  Trust me when I say that single-threaded performance as a result of clockspeed sucks!  

 

What you are referring to is IPC, in which I would agree that IPC is okay, not great.  IPC is still one notch behind Intel and clockspeed isn't even close.  

 

Single-threaded performance is a result of IPC + clockspeed.  Not just one or the other.  

 

1 minute ago, Lemon Wolf said:

Btw how did you manage the 3600 on your ram? It doesnt matter what i try i cant get higher than 3500. Anything much higher than that results in boot failure.

 

I actually got it to 3733, but couldn't get it 100% stable.  It was almost there, but not quite.  3600 MHz c15-15-15 1T was a walk in the part with SOC just over 1v and sticks running 1.41v.

 

Here's 3600 MHz passing 2000% HCI Memtest.  Definitively 100% stable.  

 

[Lemon Wolf]

3 minutes ago, done12many2 said:

 

Thanks for the break down on how it works, but I can run "fixed speed" on my Intel 7900x and score 220 + in single threaded runs (~30% higher at fixed clocks).  Trust me when I say that single-threaded performance as a result of clockspeed sucks!  

 

What you are referring to is IPC, in which I would agree that IPC is okay, not great.  IPC is still one notch behind Intel and clockspeed isn't even close.  

 

Single-threaded performance is a result of IPC + clockspeed.  Not just one or the other.  

 

 

I actually got it to 3733, but couldn't get it 100% stable.  It was almost there, but not quite.  3600 MHz c15-15-15 1T was a walk in the part with SOC just over 1v and sticks running 1.41v.

 

Here's 3600 MHz passing 2000% HCI Memtest.  Definitively 100% stable.  

 

Yes i should have made it more clear, i meant the IPC of the CPU.

When i tested my 6800k and my 1950X both at 4 Ghz it was 172 vs 167 CB.

Hmm i am running the SOC at 1,1 V and my ram at 1,45V but cant get any more than 3500.

Its quite a mystery to me.

[Pannalicour]

1 hour ago, Lemon Wolf said:

Latest result with my 1950X.

 

 

To be honest the single threaded score doesnt suck at all.

Most Intel CPUs boost to quite high clock speeds when running single threaded applications.

When you test the single threaded performance at a fixed speed on both platforms the difference is minimal.

Btw how did you manage the 3600 on your ram? It doesnt matter what i try i cant get higher than 3500. Anything much higher than that results in boot failure.

The difference if the clockspeed is the same is 9 points in cinebench r15 (intel 8 serie vs ryzen)

[bimmerman]

Was messing around with my computer a bunch yesterday and figured I'd join the party.

 

Rig:

X58 Xeon X5675, OC'd at 4.7 GHz (6 core 12 thread)

Gigabyte GA-X58A-UD3R

24GB DDR3-1600, OC'd to 2050 MHz (in triple channel)

Sapphire R9 290 Tri-X

Coolermaster Glacer 240 AIO

also in a case, with a power supply, running on an ssd

 

Results:

116.63 fps

1083 cb multicore

142 cb single core (you can see a bunch of trials of my Xeon and prior i7-950 in the screenshot)

7.61x multiplier

 

I'm pretty happy with the results if I'm honest. Bumping the memory from 1600 to 2000 made a whopping 1pt of difference, so...not bothering to OC that anymore.

 

Also attached, for grins, are the firestrike results, albeit at a 200 BCLK instead of 205 (to give the memory controller a bit of a breather). This was a slight OC on the GPU from 1000/1300 to 1100/1400, which bumped scores up by ~300 points or so. Pretty negligible.

 

Not bad for a computer from the dark ages! Probably going to upgrade GPU and be set for a while, the CPU is holding its own pretty well when OCd.

[Princess Luna]

As mentioned above here it is, seems the scores got quite higher than expected, the system is performing wonderfully and it was a great upgrade from the i7 6700

 

It is a good score right? seems outperform some stock i7 8700k you find in the internet, was it the 3200mhz memory frequency that help out? I tried enabling MCE for all core turbo of 4.6ghz but it seems all cores frequency are capping at 4.35ghz regardless, temperatures are fine the Cooler Master 212X is keeping it at cool 70Cº max temp steady with CPU-z stressing for 20~25 minutes.

 

@done12many2 , @Jumper118 while the all core score seems in order I feel the single thread performance should've been a bit higher, aruond the 190ish no? am I doing something wrong here? What do you advise that I should do? I have updated every thing Gigabyte tells me to including the last BIOS revision.

 

I suppose overclocking the memory manually will be a must? it is currently using a XMP profile, it seems to be running on single channel but I am having issue getting it on dual channel given the first dimm is blocked by the 212X Fan what gives a hassle...

 

I want to ensure good performance so any tips are welcome ^^

[stlwrx]

Here are my scores - they seem to be pretty good scores, right?

 

[Tytan64]

AMD is back! I'm so glad I didn't have to build myself another Intel rig ^.^b

[agent_x007]

New toy 

[GoogleGonnaSaveUs]

Lenovo Y520

[Swiip]

That's my low power ES Xeon, only uses a max of ~50watt and is limited to 2.7ghz.

 

[DildorTheDecent]

7 hours ago, Swiip said:

That's my low power ES Xeon, only uses a max of ~50watt and is limited to 2.7ghz.

It's also an engineering sample. Doubt that it pulls just 50W. No 12 core is that low.

 

See that it has no model number, wonder what it really is.

[Swiip]

it's realy hard downclocked, the real model would have a boost clock of 3.5ghz. its a ES from an E5-2690 v3.

I cool it whit an 1x120mm AIO radiator, whitout a fan mounted on it, only the case air from 1x60mm and 1x 80mm fan go through it.

i think it was 50w whitout boost, and by 2.7 ~65w.

 

 

[DISIFTW]

[Jumper118]

8 hours ago, DISIFTW said:

 

bit lacking in the reading department?