Intel, like AMD, are large corporations with risk of lawsuits if its not fixed permanently. Lets wait and see once the microcode is roll out, meanwhile no need to wait for the fix to arrive, just pull back everything like I said above and it will help save it sooner.

A rtx 9900 card, with all the required vrm, mosfet, and RAM

Downgrade to 9800 and sell to user at 9800 price

Then ask if they want to unlock 9900??


U meed to umlock iq. Pls

Looks like a NO.

https://www.pcmag.com/news/intel-hints-13th...ermanent-damage

Keyword : "preventative"

Not "fix"

This post has been edited by Mea Culpa: Jul 28 2024, 09:19 PM

Tell that to Intel. IQ no meed much?

Class action lawsuits incoming.

https://www.tomshardware.com/pc-components/...tability-issues

Saw this news but currently , what is the condition of crashing ? or at one point of time just BOD then cannot start pc ?

How much does it affect a company all using HP prebuild with all intel gen 13/14 ?
https://forum.lowyat.net/topic/5473031 , intel extend another 2 years?

babylon52281 So what are the option is better for user desktop ? getting an AMD or 12th gen ?

Ouch. If you're an Intel fan, look away (or look elsewhere).

They need not to fix it permanently. Their liability is the warranty period. As long as they provide replacement and they will be fine.

I will reply to both with the same answers. Basically if I were to trust the internetz then AMD would have been finished with their own SOC overvoltage burn issue as well the conspiracies from 9000series delayed launch, as well I would be with Intel and this 13/14 Gen issue.

Architectually, 13/14 RPL isnt that much different that 12 ADL, so rather if you could get a good deal for 12700/12900 then why not, for the price beats later gen i5 CPUS. So get it more for the price performance rather than any concern.

If your a regular user with no intention to OC, get the regular nonK 13/14Gen. The PL2 drop to 65W is there protect. Even if you unlock PL2 limit, just remember to disable mobo auto OC, and set PL1 & PL2 to Intel PBP limit.

If your set on OCing and you realised its why you get a Ksku then you be knowing what your doing & how to minimise the degradation effects. So its a risk you knowingly take yeah.

Intel extended warranty means squat if users are lazy to do those above & mobos continue to push those CPUS beyond limit.

As for options, AM4 is an option but it lags in power efficiency to LGA1700, and still continues to bug by more unstable platform. AM5 is still overpriced due to mobo & DDR5.

This post has been edited by babylon52281: Aug 4 2024, 01:20 AM

Im not good with detail or technical things,

But i play a game that the developer refuse to fix a bug that hackers take advantege of all the time.

In the end when hackers go extreme and crash all players during playime. My ssd suddenly died after awhile. The ssd is bundle oem nameless ssd.

Soon I change into a reliable sdd. Hacker started the crashing again. I survive countless crash daily. My ssd still running until now.

A defect is a defect no need give execuses. I came across a post that follow all safety intel settings but a recent windows update totally crash his computer, His totally piss off.

As far as concerned, X3D's efficiency is top notch, but would be too slow if user's usage pattern doesn't make use of the cache.
AM5 boards are priced decently now, DDR5 too. Would rather get a 7500F and cheap A620 (and stays with 65W) or okay-ish B650 and get possibility to run newer gen processor in the future, at mid-level budget. AM4 is still unbeatable for lower budget, i.e. limited to choose 5500/5600.

---

Regular user (Yes, it presents the majority of the users) doesn't set the PL level manually, they use default settings. Would be easier if mobo provide a preset (one-click stable? as opposed by one-click OC) to apply all of the needed changes. However by imposing this limit, it will hurt the boost by a lot, rendering the system not as responsive. I would rather Intel try to do some manufacturing revisions or stricter binning, then replace the affected users with the new stepping.



This post has been edited by kingkingyyk: Aug 4 2024, 11:12 AM

Err a windows update crash has very little to do with the CPU, so blame should be given but given to the right causes. Its the same like peeps blaming Microsoft for Crowdstrike outtage, and while the its easy to blame Windows for being too sensitive to kernel faults, the blame should be going to CS for for their buggy update.

My 2sen experience, SSD seems very sensitive to hard crashes I had a MLC drive that supposed to be more durable but it died after few Windows crash. If your game is this sensitive, I suggest run it from a HDD as it has less damages from crashes, my HDDs that was connected to that MLC drive outlasted and still in use today.

Indeed it is. That's why you will find super capacitors in enterprise level SSDs to give the controller some time to flush the data back to the NAND.

From what I read, X3D has poorer idle power and has higher power usage in light use, understandable as its a gamer CPU rather than the universal CPU that is Intel monolithic. My 12700F does ~3W in internet use, X3D i read does 3X more. Gaming per watt does beat Intels which is where Ryzen does better. If usage doesnt benefit the cache, its no better or worse than the nonX versions. Its good in its niche role tho.

Decent AM5 mobos still not as cheap as decent LGA1700 mobos tho and Intel still can leverage cheaper but speedy DDR4. DDR5 doesnt have a speed over poor latency advantage until it hits 6000MHZ which is still pricier than DDR4. And if your getting Aseries mobo it will really hamper any CPU upgradability due to poorer VRMs, I would not reco pairing with current or future X3D for example, a good starter for upgrade purpose is to go for a B650 mobo with decent VRM. Much like Intel Hmobos, AMD Amobos are meant for office use with little to no intention for CPU upgrade.

But if one has the money for a decent build, yeah go for AM5. Intel Arrowlake will be new evolution so will have issues to resolve, wait for Lunarlake at least once their chiplet tech has matured.

As for preference, I would rather Intel to put the locks on their mobo partners eagerness to push CPU beyond limits (im looking at Asus, MSI as the worst). Also Intel should have hard limits even on Ksku that doesnt allow it to be driven beyond.

Here is decent AM5 example from my lazy reco list.
Gaming PC Level 1 (USD$1000 - <RM 5000) Prices based in USD converted
CPU: AMD Ryzen 7 7600X
CPU Cooler: Thermalright Assassin X120 Refined SE
Motherboard: MSI PRO B650-S WiFi Pro / Asrock B650M-HDV
RAM: Corsair Vengeance LPX 32GB (2x16GB) DDR5 6000MHz CL36
Storage: WD Blue SN850 NVMe SSD 1TB
Graphics Card: MSI Gaming GeForce RTX 4060 Ti Ventus 8GB (16GB if budget can fit)
Alternative Graphics Card option: XFX Speedster RX 7700 XT Black Gaming
Case: Corsair 4000D AIRFLOW
PSU: Corsair RM750x, 750W, Cybenetics Gold, Fully Modular

my old pc and lappy had experienced freeze and i will just hard reset, old one is 960 evo, new one is micron SSD
both SSd survived as i keep hard reset everyday lol

pc hard crash due to psu (enermax, well used and Seasonic) issue and eventually kick a bucket

older HDD tends to last longer due to lighter load, eg single platter, my hitachi 500gb and WD 320GB still running, both already run beyond the MTFB
while my toshiba 2TB died, some older model use multiplatter tend to died as soon as the warranty expired

Let's say it is 3W vs 30W for idle, but in heavy load the situation is reversed at way larger scale. So it is heavily dependent to the usage pattern.

https://www.techspot.com/review/2801-amd-ry...x3d/#Power2-png

Algorithm in the firmware is involved. Bigger brand has better experience and performs QC with this kind of scenario.

This post has been edited by kingkingyyk: Aug 4 2024, 07:41 PM

Indeed which is why I did say gaming wise its better per watt. So it really depends on usage whether its primarily just a gaming PC or a daily regular use with some hours of gaming which in that case an Intel would use less wattage per day scenario. So YMMV.

1024kbps
We talked about X86 backwards compatibility hampering its evolution progress but here is real proof how ridiculous deep it can running software way back to the DOS days.

https://www.techspot.com/news/104118-check-...n-hardware.html

Since X86 could not be weaned off its legacy limitations, there is no option than to move onto modern CPU uarchs that has no such need ie ARM & RISC V.