[Guru3D] Review: Ryzen 5 3600XT, Ryzen 7 3800XT and Ryzen 9 3900XT. - Page 8 - Overclock.net - An Overclocking Community

Forum Jump: 

[Guru3D] Review: Ryzen 5 3600XT, Ryzen 7 3800XT and Ryzen 9 3900XT.

Thread Tools
post #71 of 76 (permalink) Old 07-08-2020, 08:19 PM
Vandelay Industries
maltamonk's Avatar
Join Date: May 2015
Location: US/UK
Posts: 1,782
Rep: 85 (Unique: 63)
The MC bundle discount is $50 on these rather than the $20 with the reg. Not that it makes them worth it, but I found it interesting
maltamonk is offline  
Sponsored Links
post #72 of 76 (permalink) Old 07-08-2020, 11:10 PM
New to Overclock.net
rdr09's Avatar
Join Date: Mar 2011
Location: From the US but lives in Africa
Posts: 20,049
Rep: 894 (Unique: 644)
Quote: Originally Posted by maltamonk View Post
The MC bundle discount is $50 on these rather than the $20 with the reg. Not that it makes them worth it, but I found it interesting
Too bad the R5 does not come with Valhalla,

By Built-in Benchmark Queen:
My 2.5 years old delidded 7980xe @ 4600mhz all core with 49ns memorylatency (4000c16 twekaed) is faster than 3950x max overclocked on water in every scenario. Pretty good for a ancient cpu
In gaming it has no chance vs my new 10900k @ 5400mhz and 4600c16 tweaked memory. ~35ns memorylatency.
2nd AMD Build
(10 items)
Hard Drive
Power Supply
Operating System
▲ hide details ▲
rdr09 is offline  
post #73 of 76 (permalink) Old 07-09-2020, 04:20 AM
icehotshot's Avatar
Join Date: Jan 2011
Posts: 2,710
Rep: 147 (Unique: 125)
Quote: Originally Posted by bl4d3runn3r View Post
First time I'm a little disappointed with AMD since 2017. This CPU wasn't necessary. *SNIP*
Well I wouldn't be, this is a great move by AMD imo. This isn't about XT cpus. They get marketing hype, new reviews to take any hype away from the Intel 10 series reviews, and all the reviewers saying buy non-XT cpus, i.e. BUY AMD.

I'm really looking forward to Zen 3. Most likely they are gonna beat Intel at their last stand, gaming. GG.

(9 items)
AMD 3900X (1900 FCLK) @ 4.4 4.4 4.5 4.4
Gigabyte X570 Aorus Pro Wifi
Zotac Nvidia GTX 1080 2088mhz/1081mv/XOC + Shunt
32GB G.Skill Flare X 3200CL14 @ 3800 Mhz 15-16-15-15
Hard Drive
Samsung 970 Evo+ 1TB
Power Supply
EVGA SuperNOVA 1000 G3
Custom Loop 3 x 240mm + 1x 480mm Rads
Thermaltake Core X9
Operating System
Win 10 Pro 64-Bit
▲ hide details ▲

icehotshot is offline  
Sponsored Links
post #74 of 76 (permalink) Old 07-22-2020, 10:37 PM
New to Overclock.net
tcclaviger's Avatar
Join Date: Mar 2012
Location: Washington State
Posts: 437
Rep: 26 (Unique: 23)
I don't think they're a bad buy necessarily, if you like to tweak your computer to get the last 1% of performance the XTs seem like the CPU for you. Performance/dollar value, yeah stay away from the XTs...

I don't believe AMD for a second on "refined process" the way they are selling it, or at least, the way people are interpreting it. I think AMD updated the oxide layer thickness to survive extra voltage for longer and used the strong CCD bin equivalent from a 3900x (CCD 1) for 3600XTs, here's why:

My mid-November manufactured 3900x (a fairly good 3900x sample) vs mid-April manufactured 3600xt (a very average sample from what I can tell).

3600xt max stable single core boosting - 4727
3900x max stable single core boosting - 4747

The boost algorithm is holding it at max speed more consistently, so for a given speed reported in software, the XT scores slightly higher than the X (5 points in R20, 1 point in R15), but the 3900x hits higher speeds ultimately and they score the same single core.

3900x static OC - 4400 @ 1.3 GET (CCD 2 limited)
3600xt static OC - 4520 @ 1.32 GET

Both hit a voltage wall at 4550 requiring north of 1.42v. They will both reach 4600, but it takes far too much voltage to stabilize safely.

Sure seems like they took a single (stronger) CCD from a 3900x slapped into a 3600xt and added a tweaked boost algorithm.

The updated boosting algorithm used on the XTs boosts like using PBO does on a 3900x.

3900x unlimited PBO limits - 4350mhz @ 1.337 GET during R20
3600xt stock - 4325mhz @ 1.37 GET during R20

3900x FIT limit in P95 - 1.27
3600XT FIT limit in P95 - 1.325

Certainly seems more like they are throwing voltage at the problem to get it to boost higher during all core work, and using a slightly better bin to ensure the single core boost is stable higher up (the 3600xt does 4600 single core at all default settings). This is also possibly why the 3800xt and 3900xt don't come with a cooler, the extra voltage may (probably imho) overwhelm the Wraith and cause performance fall off.

Last edited by tcclaviger; 07-22-2020 at 10:41 PM.
tcclaviger is offline  
post #75 of 76 (permalink) Old 07-25-2020, 08:29 AM
New to Overclock.net
mongoled's Avatar
Join Date: Oct 2011
Location: Ayia Napa, Cyprus
Posts: 810
Rep: 30 (Unique: 23)

Have you pushed the IF yet ??


SeaSonic X750 Gold 750W
MSI X370 XPOWER Gaming Titanium | Ryzen 5 3600 2016SUS BCLK @107.5625 | EDC @3
Viper Steel Series PVS416G413C9K @ 3800/1900 mhz | 1.552v | 14-15-14-12-26-38-247-16-1T
Asus GTX 780 DCII OC @ 1332/3522 mhz, 1.3v | Samsung 960 EVO 500GB
EK-FB MSI X370 XPower RGB Monoblock 1/2" ID | Swiftech D5 MCP655 | EK-XRES 100
Alphacool NexXxoS UT60 280 | 4 x Scythe GT AP-15 @ 5v
Thermochill PA120.3 | 1 x NF-A12x15 | 2 x ML120 | PWM
mongoled is offline  
post #76 of 76 (permalink) Old 07-26-2020, 08:16 AM
New to Overclock.net
tcclaviger's Avatar
Join Date: Mar 2012
Location: Washington State
Posts: 437
Rep: 26 (Unique: 23)
1904ish is the limit on my 3600xt, 1907 on 3900X.

Just like my 3900x, for 1900IF, 100 bclk is not stable with the 19.00 IF strap, but 100.2 and 19 is stable, without adjusting any voltages. I blame Spread Spectrum on the C7H, it's enabled at 100 bclk, and can't be turned off.

I posit this is why many people can't get 1900 stable, SS is terrible and Asus should feel terrible for not allowing it to be switched off.

Between the two, 1904/3808 at the same RAM settings the 3900x gives a little better read speeds and about 1.5ns lower latency.

My 3900x is not the greatest to compare this 3600xt with, great chip vs average.

I found 103.6 to be a key number for bclk and aligning boosting behavior with IF clock. 101.8 or 102 next best choice imo if it won't stabilize at 103.6.

46.25x 103.6 = 4800mhz single boost
45.25x 103.6 = 4688mhz all core peak boost
18.33x 103.6 = 1899 IF

The XT is drastically more aggressive with all-core voltages and surprisingly more conservative with single core voltage.

Last edited by tcclaviger; 07-26-2020 at 08:21 AM.
tcclaviger is offline  

Quick Reply

Register Now

In order to be able to post messages on the Overclock.net - An Overclocking Community forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.
User Name:
If you do not want to register, fill this field only and the name will be used as user name for your post.
Please enter a password for your user account. Note that passwords are case-sensitive.
Confirm Password:
Email Address
Please enter a valid email address for yourself.
Email Address:


Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page

Forum Jump: 

Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off