Overclock.net banner

1 - 20 of 202 Posts

·
Premium Member
Joined
·
1,159 Posts
Discussion Starter #1 (Edited)
Let's collect and discuss all the info we can gather about the VRMs on boards with Z370 or Z390 PCHs Z490 PCH in this thread. (br0da 2017)


V3 includes some vrm temp results as requested 13/6/20
V2.5 RELEASE PRICES ON NEWEGG US INCLUDED 6/5/20
V2 VRM READY 4/5/20
V1 START 30/4/20



Due to limited spaces here, only vcore info will be listed below.
check out the google doc for more details like vccgt/vccsa/vccio/vdimm(ddr)/pcb layer/ddr4 layout/sources of reviews and pic for proof/release prices on newegg.
https://docs.google.com/spreadsheets/d/16YJm4L1-ohpL8s-4rLDDDCBZvi97ZYwkc44s7LS5-2Q/edit?usp=sharing
you are wellcome to provide links/pics here to improve the accuracy of vrm data together, really appreciate that.
?=i dont know/not sure


For simiplicity,
teamed=
intergrated drivers for each and every mosfets(drmos/smart power stage has driver builtin)
external discrete drivers for each 1H+1L like 4C09N+4C06N (single n channel discrete fet), and for each dual n channel fet like CSD87350 (combining H+L in one package).

parallel=
two pairs of 1H+1L discrete fets [=2*(1H+1L)] or two dual n fets with two chokes share one driver

----------------------------

VCORE VRM INFO:
(check the excel for latest update)

ASROCK

VCORE + VCCGT + VCCSA + VCCIO, VCORE CONTORLLER, VCORE MOSFET, VCORE DESIGN

Z490M-ITX/ac
5?+2?+1?+1?, UP9521R (5+2?), 5*SIC654 50A = 250A?, DIRECT?

Z490M Pro4
8+2+1+1, UP9521R (4+2), 8*(1H+1L) PDEC3908 + PDEC3906, DOUBLING 4*UP1961S

Z490 PRO 4
8+2+1+1, UP9521R (4+2), 8*(1H+1L) SM4337+SM4336, DOUBLING 4*UP1961S

Z490 Phantom Gaming-ITX/TB3
6?+2?+1+1?, ISL69269 (6+2+1?), 6*ISL99390 90A = 540A, DIRECT

Z490 Phantom Gaming 4 (2.5G/AX/AC)
8+2+1+1, UP9521R (4+2), 8*(1H+1L) PDEC3908 + PDEC3906, DOUBLING 4*UP1961S

Z490 PG Velocita
10+2+1+2, ISL69269 (10+1+1), 10*SIC654A/SIC632A 50A = 500A, DIRECT

Z490 Steel Legend
8+2+1+1?, ISL69269 (8+2+1), 8*SIC654A//SiC632A 50A = 400A, DIRECT

Z490 Extreme4
8+2+1+1?, ISL69269 (8+2+1), 8*SIC632A 50A = 400A, DIRECT

Z490 TAICHI
12+2+1+2, ISL69296 (6+2+1), 12*SIC654 50A = 600A, DOUBLING 6*ISL6617A

Z490 AQUA
90A








GIGABYTE

VCORE + VCCGT + VCCSA + VCCIO, VCORE CONTORLLER, VCORE MOSFET, VCORE DESIGN

Z490M
8+2+1+2?, ISL (4+2)???, 8*(1H+1L) 4C10N+4C06N?, 4*driver?, PARALLEL
(parallel 8 or direct 4, with 4 chokes and 8H+8L and probably 4 drivers in total)

Z490 GAMING X
11+1+1+2?, ISL69269 (11+1+0), 11*SIC651A 50A = 550A, DIRECT

Z490 UD (AC)
11+1+1+2?, ISL69629 (11+1+0), 11*(1H+1L) 4C10N+4C06N, 11*DRIVER?, DIRECT

Z490M GAMING X
11+1+1+2?, ISL69629 (11+1+0), 11*(1H+1L) 4C10N+4C06N?, 11*DRIVER?, DIRECT

Z490 VISION D
12+1+1+2, ISL69296 (12+0+0), 12*SIC651A 50A = 600A, DIRECT

Z490 VISION G
12+1+1+2, ISL69296 (12+0+0), 12*SIC651A 50A = 600A, DIRECT

Z490I AORUS ULTRA
8+1+1+2, ISL69296 (8+1+1), 8*ISL99390 90A = 720A, DIRECT

Z490 AORUS ELITE (AC)
12+1+1+2, ISL69296 (12+0+0), 12*SIC651A 50A = 600A, DIRECT

Z490 AORUS PRO AX
12+1+1+2, ISL69296 (12+0+0), 12*SIC620A 60A = 720A, DIRECT
(gigabyte on its official site claims aorus pro ax has "55A DrMOS" used. but the reviews shows the drmos is SIC620A rated at 60A.
on SIC620A specs sheet, the graphs show at 125c Tpcb + 300khz, SIC620A=55A ; (100C=60A))


GZ490 AORUS ULTRA (G2)
12+1+1+2, ISL69296 (12+0+0), 12*SIC620A 60A = 720A, DIRECT

Z490 AORUS MASTER
14+1+1+2, ISL69269 (7+1+1), 14*ISL99390 90A = 1260A, DOUBLING 7*ISL6617A

Z490 AORUS XTREME (WATERFORCE)
16+1+1+2, ISL69269 (8+1+1), 16*ISL99390 90A = 1440A, DOUBLING 8*ISL6617A






ASUS

VCORE + VCCGT + VCCSA + VCCIO, VCORE CONTORLLER, VCORE MOSFET, VCORE DESIGN


PRIME Z490M PLUS
8+1+1+1, ASP1900B (4+1), 8*SIC639 50A = 400A, TEAMED

PRIME Z490 V
8+1+1+1, ?, 8*DRMOS, TEAMED

PRIME Z490 P
10+1+1+1, ASP1900B (5+1), 10*NCP302045 45A = 450A, TEAMED

PRIME Z490 A
12+2+2+2, ASP1900B (6+2), 12*NCP302045 45A = 540A, TEAMED

PROART Z490 CREATOR 10G
12+2+2+1, ASP1900B (6+2), 12*NCP302045 45A = 540A, TEAMED

TUF GAMING Z490 PLUS (WIFI)
12+2+2+1, ASP1900B (6+2), 12*SIC639 50A = 600A, TEAMED

ROG STRIX Z490 H GAMING
12+2+2+2, ?, 12*DRMOS, TEAMED

ROG STRIX Z490 I GAMING
8+2+2+2, ASP1405I (4+2), 8*TDA21462 60A = 480A, TEAMED

ROG STRIX Z490 G GAMING (WIFI)
12+2+2+2, ?, 12*DRMOS, TEAMED

ROG STRIX Z490 F GAMING
12+2+2+2, ASP1900B (6+2), 12*NCP302045 45A = 540A , TEAMED

ROG STRIX Z490 E GAMING
14+2+2+2, ASP1900B (7+1), 14*SIC639 50A = 700A, TEAMED

ROG STRIX Z490 A GAMING
12+2+2+2, ASP1900B (6+2), 12*NCP302045 45A = 540A, TEAMED

ROG MAXIMUS XII HERO (WI-FI)
14+2+2+2, ASP1405I (7+1), 14*TDA21462 60A = 840A, TEAMED

ROG MAXIMUS XII FORMULA
16+0+2+2, ASP1405I (8+0), 16*TDA21472 70A = 1120A, TEAMED

ROG MAXIMUS XII EXTREME
16+0+2+2, ASP1405I (8+0), 16*TDA21490 90A = 1440A, TEAMED

ROG MAXIMUS XII APEX
16+0+2+2, ASP1405I (8+0), 16*TDA21472 70A = 1120A, TEAMED




MSI

VCORE + VCCGT + VCCSA + VCCIO, VCORE CONTORLLER, VCORE MOSFET, VCORE DESIGN


MEG Z490 GODLIKE
16+1+1+2, ISL69269 (8+1+1), 16*ISL99390B 90A = 1440A, DOUBLING 8*ISL6617A

MEG Z490 ACE
16+0+1+2, ISL69269 (8+0+1), 16*ISL99390B 90A = 1440A, DOUBLING 8*ISL6617A

MEG Z490 UNIFY
16+0+1+2, ISL69269 (8+0+1), 16*ISL99390B 90A = 1440A, DOUBLING 8*ISL6617A

MPG Z490 GAMING CARBON WIFI
12+1+1+2, RAA229001 (6+1), 12*ISL99360 60A = 720A, PARALLEL

MPG Z490 GAMING EDGE WIFI
12+1+1+2, RAA229001 (6+1), 12*ISL99360 60A = 720A, PARALLEL

MPG Z490M GAMING EDGE WIFI
12+1+1+2, RT3609BE (6+1), 12*(1H+1L) 4C029+4C024 , 6*RT9624F (4P=) driver??? PARALLEL (unknown drivers)

MPG Z490 GAMING PLUS
12+1+1+2, RT3609BE (6+1) 12*(1H+1L) 4C029+4C024 , 6*RT9624F (4P=) driver, PARALLEL

Z490 A PRO
12+1+1+2, RT3609BE (6+1), 12*(1H+1L) 4C029+4C024, 6*RT9624F (4P=) driver??? PARALLEL (unknown drivers)

MAG Z490 TOMAHAWK
12+1+1+2, NCP81229 (6+1), 12*NCP302155 55A = 660A, PARALLEL

MEG Z490I UNIFY
8+1+1+2, ISL69269 (8+1+1), 8*ISL99390B 90A = 720A, DIRECT







BIOSTAR

VCORE + VCCGT + VCCSA + VCCIO, VCORE CONTORLLER, VCORE MOSFET, VCORE DESIGN


Z490GTN Ver. 5.0
6+2+1?+1?, NCP81229 (6+1), 6*NCP302155 55A = 330A, DIRECT

Z490GTA Ver. 5.0
12+2+1+1, NCP81229 (6+1), 12*(1H+1L) 4C029+4C028, 12*NCP81258 DRIVER, DOUBLING 6*NCP81162

Z490GTA EVO Ver. 5.0
14+2+1+1, NCP81229 (7+1), 14*(1H+1L) 4C029+4C028, 14*NCP81258 DRIVER, DOUBLING 7*NCP81162




SUPERMICRO

VCORE + VCCGT + VCCSA + VCCIO, VCORE CONTORLLER, VCORE MOSFET, VCORE DESIGN


C9Z490-PGW
6+2+1+1, XDPE12284C (6+2), 6*TDA21490 90A = 540A, DIRECT





EVGA

VCORE + VCCGT + VCCSA + VCCIO, VCORE CONTORLLER, VCORE MOSFET, VCORE DESIGN


Z490 DARK
12+2+1+1, ?, 12*ISL99227 60A = 720A, ???



COLORFUL

VCORE + VCCGT + VCCSA + VCCIO, VCORE CONTORLLER, VCORE MOSFET, VCORE DESIGN

iGame Z490 Vulcan X V20
12+2+1+1?, RAA229001 (6+2), 12*SM7341EHKP, DOUBLING 6*ISL6617A

CVN Z490 GAMING PRO V20
8+2+1+1?, NCP81228 (8+2), 8*SM7341EHKP , DIRECT


---------------------------------------

PLEASE READ THE BELOW GIGABYTE CHEATSHEET CAREFULLY!
Gigabyte z490 VRM CHEATSHEET, provided by Gigabyte official on reddit
https://docs.google.com/spreadsheets/d/15jArPp1FvQL7wnXeaVFE_9_PINNFFT8gmvcI3NTc5F4/edit#gid=0
after a lot of researching, i found out the data on that cheatsheet may not be accurate.






NOTES:
1. Z490 is very likely to support 11th gen rocket lake cpus.

2. Due to the new Intel spec for 11th gen rocket lake cpu (for upcoming feature support), lots of Z490 have been designed to have 2 phases of VCCIO.
-->New VCCSA design is changed to SVID, just like VCORE and VCCGT.
-->New VRM CONTROLLER is needed to controll 3*SVID (=VCORE+VCCGT+VCCSA), eg INTERSIL ISL69269 12 phases controller(X+Y+Z=12).
(this is not strictly required by Intel, mobo vendors are free to use the current way to control VCCSA and IO on Z390)
check the attached pic for more new Intel Spec for 11th gen rocket lake. https://www.techbang.com/posts/78199-intels-10th-gen-core-desktop-processor-comet-lake-s-is-here-400-chipset-compatible-with-rocket-lake-s?from=home_headline

3. VCCGT is worth to check out as the intel new spec for 11th gen rocket lake is 55a. while some models of z490 have 1 signle phase for vccgt and it is rated at 50a only.

4. ASUS TEAM vrm design differs from traditional parallel vrm, check the attached pic by (ASUS ROG R&D BING)
in short, teamed design requires drivers for each 1h+1l/dual n (or drmos/smart power stage) so that each 1h+1l/dual n (or drmos/smart power stage) has its own circuit and each of two 1h+1l/dual n/drmos/sps need not to be wired together to be driven by the same driver.





SOME NEW FEATURES OTHER THAN VRM ON Z490:
i) NO GEN4 SUPPORT ON Z490 WITH 10TH GEN CPU.

ii) Z490 LGA1200, socket size is compactible with LGA 115X meaning current LGA 115X cooling mounting is compactible on Z490 too.

III) Z490 is very likely to support 11th gen rocket lake cpus which have native gen4. (confirmed by GIGABYTE OFFICIALS)

IV) Lots of Z490s are PCIE4.0 ready on the first M.2 slot and the first X16 PCIE slot. asus has m.2 gen4 ready on m12e only, for the first pcie slot check the below pic (pcb quality of prime plus & prime p may not meet the gen4 standard.)

V) 9900K die height (without STIM) is 0.08mm; 10900K die height (without STIM) is 0.04mm (same as 8700k/8086k).
The thickness of IHS of 10900K is 0.4mm; The thickness of IHS of 9900K is 0.6mm.

VI) intel confirms that there are bugs in its 2.5g lan I225. and intel later responds I225V2 with bug fixed is in production already(check below attachment).
while according to some reliable source, it is also said that even B2/V2 has some bugs that have to be fixed on B3, thats why some mobos simply skip the intel 2.5g and put intel 1g lan or realtek latest 2.5g RTL8125B (package loss on previous verison RTL8125 is fixed).
in short, B1=SLN9C & SLN9D; B2=SLNJY.

asus has both versions of V1/B1 and V2/B2 used on its z490s at least on review samples:
https://www.overclock.net/forum/28451958-post95.html

but asus china officials said retail version would be come with v2/b2.
https://www.overclock.net/forum/6-intel-motherboards/1746916-1440a-madness-z490-vrm-discussion-thread-v2-5-6-5-20-a-10.html#post28451958

intel's response to wccftech report confirming B2/V2 has the bug fixed
https://wccftech.com/intel-400-series-chipsets-z490-i225-v-network-controller-issue-fix-2h-2020/

18-5-20 update:
from another source, it is said that gigabyte would be using the latest intel i225v2 (b2) chips on its retail version's z490s.

==>therefore if asus and gigabyte have the latest b2/v2 chips used then there should then be no problem according to intel's claims.

23-5-20 update:
there is report that even b2/v2 of i225 intel 2.5g lan has bugs....
www.wccftech.com/intel-foxville-i225-v-2-5gbe-networking-issues-persist-z490-motherboards-affected/



VII) For realtek 2.5g lan, it is also said that the the RTL8125 (eg on x570) has packet loss problem. while this time some z490 mobo vendos have opted for the latest version of realtek 2.5 lan called RTL8125"B" which is claimed to be packet loss fixed. so check the exact model of realtek 2.5g lan if you care about this problem(RTL8125B).

VIII) z490 itx concern---eps cpu connectors on z490:
some mobos (asrock, asus, gigabyte) have so-called solid pin design on z490. while asus this time on its z490 promotes procool II (solid pin) to have 10a on each 12v of 8 pin connector(with 4*12v). therefore, 10A*4=40A for each asus eps connector with procool II. what you should worry about is the quality of your psu's eps 8 pin headers. even solid pin design cant keep you safe if your psu's eps 8 pin headers suck. check the attached pics and that mobo is in solid pin design.
for other mobo vendors, you should be expected to get ~7a each on the mobo 8 pin, giving you around 340w headroom. dont need to worry too much on mobo side, as your psu cables headers may burn first. custom cables emphasises a lot on the quality of wires, but headers matter too.
current passes throught the surface of the pin, not the solid pin inside, thats why some mobo vendor still dont buy it. while solid pin design is bad for ppl who would keep plugging and unplugging as the solid pin would be damaged by that. hallow pin has more flexibility for that.
tldr-->watch out! it is time to test your third party custom cables lol. (attached pics come from internet, not mine)
according to greentechreviews, single 8pin eps connector handling ~300w for a hour is expected to be 90~100c, eps cables near the connector ~90c.

IV) some gigabyte low end models like vision g has three m.2 slots. however one is locked for futher cpu, meaning you cant use that right now with 10th gen cpus and if the dreams never come true then that is the slot you can never use. the reason i guess should be related the wiring design and the absense of switches. gigabyte seems to have that particular gen 4 m.2 slot to be wired to cpu directly only and have not also wire it to chipset unlike its high end models like master. for the gen 4 m.2 slot to work right now as pcie3.0x4 with 10th gen cpu, it has to be wired to chipset as well, and with pcie gen 4 switches for it to take pcie lanes from chipset with 10th gen cpu right now, while switch it back to cpu lanes when 11th gen cpus arrived.
some vendor has all its z490s' first m.2 slot (close to cpu socket) to be gen 4 ready and wire it not just to cpu but also to chipset as well with gen4 switches on pcb, therefore all two m.2 slots (low end models) can be used as pcie3.0x4 right now with 10th gen cpus. gigabyte simply offers an additional m.2 slot for 11th cpu sepecific, you can still have two m.2 slots to be used as pcie3.0x4 right now with 10th gen cpus. but i am not sure whether when 11th gen cpus arrived, all three m.2 slots could be used at the same time (at different pcie speed) on gigabyte low end mobos, we will see.
tldr, you must read the guide!
check the attached pics. gigabyte aorus elite has only two m.2 slots wired to pch vs msi godlikd has shown the gen4 m.2 complete wiring with switches (the mystic red line!)
https://www.gigabyte.com/Motherboard/Z490-VISION-G-rev-10/support#support-manual


26/7/20 update
add one picture of asu strix z490-i @300w 10900k 5.3ghz 30mins blender
->110c on cpu eps cable.
https://greentechreviews.ru/2020/07/22/obzor-materinskoj-platy-asus-rog-strix-z490-i-gaming/3/



special credits (in no particular order):

gtrlab
From GREENTECH_REVIEWS, special thanks for providing high resolution pics

GBT_Brian
From Reddit, GIGABYTE VRM CHEATSHEET provided

UNIKOSHARDWARE
From Taiwan

ENTERBOX
From Taiwan

TOPPC LIN
FROM MSI, PROVIDING MAIN VRM INFO OF MSI Z490

R.F.
FROM TECHBANG TW, confirmed two phases of vccsa on asus strix a is teamed rather than direct-->asus vccsa 1 phases with 2 power stages/mosfets and chokes

Falkentyne
From overclock.net, sharing overclocking experiences on z390 aorus master, a mobo has almost lower vrm temp among z390s.


















Previous thread for z370 and z390
https://www.overclock.net/forum/6-intel-motherboards/1638955-z370-z390-vrm-discussion-thread.html

i really want to emphasise the vrm info above is not meant to be directly compared. you can never judge a vrm design by looking above data.
what we have learnt on the previous thread (above link) of z370 & z390 is that mosfets on pcb could be very strong but in reality it sucks.
the best example comes from gigabyte z370 aorus 7, with 8 phases ISL 60A smart power stages by 4 doublers, probably the best z370 on pcb. and it is reported by various users on the thread that the vrm temp would goes up to 120c, for 8700k 5ghz. even with the mod/fix later on, it is still ~90c. and this mobo has a little vrm fan built in.
the whole performance of vrm matters, not mosfets only. cooling plays a substantial part. current rating of mosfet in spec sheet could be at peak only, so check the corresponding loading temperature, eg 60a rated at 25c means nothing to mobos in reality.

---------------------------------------
**********
VRM temps are not everything
https://www.overclock.net/forum/28441184-post70.html
by Falkentyne

Z390 Master had excellent VRM temps. But due to some bug, overclocking required higher voltage if VRM switching frequency was higher than 300 khz. 300->500 khz required about 15-20mv higher vcore for the same stability at LLC:Turbo (0.4 mohms). Maybe this was a problem with the IR 35201, but I don't own an oscilloscope to check this. This issue does not happen if Loadline Calibration is left at Intel default spec (1.6 mOhms = "Standard / Normal"). So unless you are using the highest vdroop LLC, Switching frequency must be set to 300 khz for the most OC headroom.

I do not know if this a BIOS bug or an issue with the IR 35201 when vdroop is reduced. I do not have access to an oscilloscope, or extra hardware. I would need access to other boards with the same VRM controller, like MSI Z390 Meg Ace or Asus (ASP? IR rebrand?) Z390 Apex or Extreme. An oscilloscope would at least tell me what's happening with the minimum voltage transients.

So no, VRM temps are not everything. They are probably not even important at all unless you exceed 85C. There are so many other things that affect transient response (ripple), like capacitor layout, as shown on the Z390 Dark. It's a shame that buidzoid was never able to get an oscilloscope capture of the XI Apex. But the 6 phase twin phase Gene didn't do poorly at all, it did nicely vs the Master, but the Dark destroyed everything.
*****************
--------------------------------------------------------------------------------------------------------------------------------------------------------------------------
 

Attachments

·
Registered
Joined
·
1,130 Posts
what happened? It seems most Z490 pwm have shifted to ISL doublers from IR direct!

Many Z490 have active cooling on VRM too! Is this the reason? ISL have higher current VRM?
 

·
Registered
Joined
·
1,810 Posts
could we also include which boards use daisy chain/t-topology for mem path?
 

·
Premium Member
Joined
·
1,159 Posts
Discussion Starter #5
what happened? It seems most Z490 pwm have shifted to ISL doublers from IR direct!

Many Z490 have active cooling on VRM too! Is this the reason? ISL have higher current VRM?
it is for the 11th gen cpu requiring a different way to control VCCSA. The new ISL controller support X+Y+Z=12, instead of old current controller supporting X+Y=12 only. combining vccsa into vcore+vccgt may result in higher efficiency on the new isl controller according to msi. however asus still uses their own asp controller having X+Y=8 only.


11th gen cpus require better vrm and more phases for vccgt and vccsa and vccio.
 

·
Premium Member
Joined
·
1,159 Posts
Discussion Starter #6
could we also include which boards use daisy chain/t-topology for mem path?
sure, as well as the pcie bclk oc part!
and i am dying!

it is hard to confirm the layout of ddr4 wiring by reviewers. without exact words from officials we can only guess that, just like pcb layers.
i dont know whether i should include everything here, as this used to be a vrm thread lol.

needs you help to find more sources of reviews (with pics) to confirm the design of z490.
 

·
Registered
Joined
·
11 Posts

·
Premium Member
Joined
·
1,159 Posts
Discussion Starter #9
amazing qaulity, nice previews, looking forward to see your testing results!
i will put your name in the chart:D

the vccsa and vccio parts really get me...
 

·
Registered
Joined
·
490 Posts
what about XII Apex?
 

·
Registered
Joined
·
1,135 Posts
Apex is usually the best mobo in the world for extreme memory OC and CPU, I think this time won't be different, right?
 

·
Registered
Joined
·
181 Posts
Can we also include PCI-E 4.0 'hardware ready' ? This is important as it appears some boards aren't going to support it or don't have the necessary hardware.

So far it seems MSI, AsRock, and Gigabyte boards support PCIE 4.0 hardware and will support it with Rocket Lake CPUs. EVGA and Asus both have no mention of it in their promotional materials or under specs on their product web page.

This is a pretty big deal if you don't want to purchase a new motherboard down the road just to get PCIE 4.0 which will likely be supported by all new graphics cards going forward.
 

·
Registered
Joined
·
11 Posts
EVGA and Asus both have no mention of it in their promotional materials or under specs on their product web page.
Because it's under NDA. Currently Intel is preparing their official statement about this situation.

So far it seems MSI, AsRock, and Gigabyte boards support PCIE 4.0 hardware and will support it with Rocket Lake CPUs
And they violate the NDA with their statements about PCIE 4.0.
 

·
Premium Member
Joined
·
1,159 Posts
Discussion Starter #17 (Edited)
Can we also include PCI-E 4.0 'hardware ready' ? This is important as it appears some boards aren't going to support it or don't have the necessary hardware.

So far it seems MSI, AsRock, and Gigabyte boards support PCIE 4.0 hardware and will support it with Rocket Lake CPUs. EVGA and Asus both have no mention of it in their promotional materials or under specs on their product web page.

This is a pretty big deal if you don't want to purchase a new motherboard down the road just to get PCIE 4.0 which will likely be supported by all new graphics cards going forward.
it is not hard to check whether the model of z490 does support gen 4 or not.
gen 4 ready in hardware level means there are gen 4 specific hardwares/chips on pcb, like pcie gen4 switches, and gen4 bclk generator.

for example, gen 4 switches are needed for the (first) m.2 slot to get back the 11th gen rocket lake cpu's storage specific gen4 x4 lanes.
current z490 m.2 slots are all wired to z490 chipset, and lots of z490s have also wired to cpu to be gen 4 ready later.
 

·
Registered
Joined
·
492 Posts
Because it's under NDA. Currently Intel is preparing their official statement about this situation.



And they violate the NDA with their statements about PCIE 4.0.
You could build an old 2011 socket board with PCIE 4.0 parts on it and call it PCIE 4.0 ready, I don't see brands on their board pages talking about next generation CPUs with PCIE 4.0 at all. Point that out where they do as I just looked at a few and did not see it.

Thanks for the pictures!
 
1 - 20 of 202 Posts
Top