Jump to content

User talk:Rando717

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia


Welcome!

[edit]

Hi Rando717! I noticed your contributions to Graphics card and wanted to welcome you to the Wikipedia community. I hope you like it here and decide to stay.

As you get started, you may find this short tutorial helpful:

Learn more about editing

Alternatively, the contributing to Wikipedia page covers the same topics.

If you have any questions, we have a friendly space where experienced editors can help you here:

Get help at the Teahouse

If you are not sure where to help out, you can find a task here:

Volunteer at the Task Center

Happy editing! ― Blaze WolfTalkBlaze Wolf#6545 19:12, 22 October 2022 (UTC)[reply]

ArbCom 2022 Elections voter message

[edit]

Hello! Voting in the 2022 Arbitration Committee elections is now open until 23:59 (UTC) on Monday, 12 December 2022. All eligible users are allowed to vote. Users with alternate accounts may only vote once.

The Arbitration Committee is the panel of editors responsible for conducting the Wikipedia arbitration process. It has the authority to impose binding solutions to disputes between editors, primarily for serious conduct disputes the community has been unable to resolve. This includes the authority to impose site bans, topic bans, editing restrictions, and other measures needed to maintain our editing environment. The arbitration policy describes the Committee's roles and responsibilities in greater detail.

If you wish to participate in the 2022 election, please review the candidates and submit your choices on the voting page. If you no longer wish to receive these messages, you may add {{NoACEMM}} to your user talk page. MediaWiki message delivery (talk) 01:55, 29 November 2022 (UTC)[reply]

A few words of advice

[edit]

Hello, it's AP 499D25!

Thank you for your presence at, and contributions to Wikipedia.

I just wanted to point out a few tips and things that you may not know about.

First one, is about editing talk page content. Generally, other peoples' talk page messages should not be edited unless it's significantly off-topic or contains personal attacks or if there's any major reasons like that. This WP guideline section goes over some valid and reasonable changes that can be made to others' comments, as well as best practices for doing so. I don't mind at all that edit you made to AMD Radeon Software talk page, but I still wanted to point this out. You may have noticed a bot has unintentionally added an "unsigned" signature as a result of the edit, which is one of little consequences that can arise from doing so.

Next up, is about removing vandalism, unconstructive edits, link spam, that kind of stuff. You may want to consider posting a warning on the user or IP's talk page, using one of the appropriate warning templates listed here, to let them know that what they have done is unconstructive or goes the rules and guidelines, next time. It also helps admins know about a user's previous conduct before blocking them. There are even WP gadgets and scripts like Twinkle that can semi-automate the task for you, and make things easier and quicker. Though, for the seemingly more experienced/knowledgeable editors, I recommend using good judgment in whether to post a personally written notice, or template them.

Thirdly, wikilinks. There is actually no need to obsess over wikilinks and whether to un-redirect or not. Sometimes it's good to do such a thing, but there are also times when it is unnecessary. One example of unnecessary link change is 7 nm --> 7 nm process|7nm. There is no need to place a pipe and target if the link works without it, like here: 7 nm. It only should be done for if it results in a red link otherwise. Even I myself made this mistake a few times in the past, not knowing about WP:NOTBROKEN. The kind of redirects and links that should be fixed however, are double redirects, since these actually harm the user experience as they stop after the first redirect and don't go through the second.

Last up, this is not related to or adds to any of the above. But it is a quick notice regarding what's been going on in the Ryzen 7k series CPU template. The next time someone reverts the table without any achieved consensus or talk page discussion, I would advise you not to revert it again, instead I'm going to request temporary semi-protection for it, which is a good solution for repeated disruptive edits by multiple suspected sockpuppet IPs.

Happy editing, and happy new year in advance!

AP 499D25 (talk) 03:39, 29 December 2022 (UTC)[reply]

Hey @AP 499D25
I don't know what to tell you about that talk page.A honest mistake?
After I replied reflist was auto generated below it, it looked odd so without any thinking I added reflist.
But than bot instantly made it worse by signing and copying entire message again.
There was no way for me to fix it or revert 2 edits and hope that bot won't sign it again.
Looking at that WP guideline; if I added {Reflist-talk} (as part of fixing format errors) instead of <references> it would be fine?

About vandalism, tbh I don't really wanna get involved going back and forth with annon IPs and risk my account.
I am perfectly fine doing stuff manually (no need for scripts/bots).If it is same page/article; I gonna revert it "once a day" and that's it.

I didn't have need to create redirect pages, so naturally I didn't look into redirect guidelines.I was kinda aware of WP:NOTBROKEN, but never looked into it.
To me it still makes more sense linking to exact section insted of using redirect.
For example [DirectX#DirectX 12|DirectX 12] instead of [Directx12]. I know that sections are more prone to changes than article titles (so are the redirects if section is removed/renamed right?).
Usually I followed what others do and kinda pickup on it.

About recent template reverts: Are you aware that it's the same person that got blocked like 3 times already (@LeaveMeB, @CristoCalis, @Halvleder)?
I am no snitch but he has new account yet again (way too obvious behaviour, I am sure you can figure it out),
except this time he decided to be more "stealthy" by login off and doing annon IP reverts.
Still same old harassment and sockpuppetry. Rando717 (talk) 16:56, 30 December 2022 (UTC)[reply]
Hello Rando717,
I see, that's understandable. I may try out the reflist-talk template on a talk page that could benefit from it soon, and see how it goes. The bot probably isn't very "aggressive" so it likely won't put unsigned comment again when it is removed.
---------------------------------------------
There is a three revert rule that states that an editor must not revert the same change more than three times in a 24-hour period. However, it also states that some edit reversions are completely exempt from the rule, such as vandalism or edits by block-evaders.
---------------------------------------------
I see, that makes sense too.
Sometimes I may edit the redirect itself to point to the precise part of the target page if it doesn't already, or is out of date. For example socket sWRX8 used to just redirect to Socket SP3 article in general and not point to anything specific, so I edited it to point to Socket SP3#SP3r4 which is the relevant part of that page covering sWRX8.
I use the visual editor quite a lot and it has a bit of a limitation on the freedom I have when it comes to making links. Which I found out it is where some imperfect links come from.
Yes, when I started editing Wikipedia some of the things I learned were from looking at what other editors do.
---------------------------------------------
Yes I'm aware it appears to be the same person who was behind those sock accounts.
What I was trying to do was to reach out to them to start discussion and give them a bit of a 'last chance' to form consensus and collaborate, before going all out on reporting them to admin noticeboard and requesting semi-protection on the Ryzen 7k template. However you may or may not have noticed, another editor has done all that already. Which I'm fine and cool with, as trying to "force" consensus through repeated disruptive edits is quite unacceptable, both in my opinion and according to WP rules.
Earlier on, someone did however, write a lengthy comment on the linked talk page expressing their full thoughts and criticism of the table design.
AP 499D25 (talk) 08:25, 31 December 2022 (UTC)[reply]
"Earlier on, someone did however, write a lengthy comment on the linked talk page..."
That's good, by communicating, the way it should be, but the key word here is someone.
Sock accounts opinion hold no value to me, and I don't see point having any further discussions with said editor.
(quick comparison of Special:Contributions should reveal enough, I just don't have time to deal with WP:SPI)
----------
sidenote: Same anonymous editor added branding rows to Intel Arc templates (you can check edit history and compare with WHOIS/Geolocate) Rando717 (talk) 18:07, 31 December 2022 (UTC)[reply]
@Rando717 Can you temporarily enable your email? Or maybe send an email to me, after which I will follow up with what I wanted to ask? AP 499D25 (talk) 07:33, 2 January 2023 (UTC)[reply]
Hi, could you check your email again (the one you used last time)? I have sent something here. — AP 499D25 (talk) 06:58, 26 March 2024 (UTC)[reply]
@AP 499D25 Done. Rando717 (talk) 08:12, 26 March 2024 (UTC)[reply]

Retirement

[edit]

Sad to see you go!

I look at your contributions and notice all the great, wonderful edits that you made over the previous many months, properly formatting citations, fixing infobox code, general cleanup, etc. These kinds of edits are invaluable to Wikipedia! And there's not many editors like you out there.

Don't let the words of advice I've left above discourage you! They are some recommendations after all.

I'm also here to report hey, that sockpuppet on AMD and Intel articles has finally seemingly gone away for good after the latest account block.

With a bit of strength and courage, comes victory!

I have also started doing some recent changes patrolling over the last couple of days, looking at new edits and seeing if they're vandalism, warning the user on their talk page, and filing a report at admins' noticeboard when the vandalism persists.

I hope to see you return when you feel like editing again.

Cheers!

AP 499D25 (talk) 12:40, 24 January 2023 (UTC)[reply]

A barnstar for you!

[edit]
The Original Barnstar
This barnstar is for all the great edits you have made on Wikipedia during your time on here: fixing citations, redirects, general cleaning up of articles, and more! There's not many other editors like you out there who make significant improvements to the quality and consistent feel of articles. Thank you for coming here! AP 499D25 (talk) 12:55, 24 January 2023 (UTC)[reply]

Thanks again!

[edit]

Hi Rando717, I want to thank you for one more excellent contrib. to Transistor count (https://en.m.wikipedia.org/wiki/Special:MobileDiff/1145828250), in particular calling into question the GA106 number – I fixed it both there and in GeForce 30 series after seeing your remark. If you spot something else, don't hesitate to mention it or just fix yourself (the GA106 count was an obvious error judging by density and TPU database (which does contain erroneous data though), unlike the more insidious Navi31 issue I recall discussing at length on RDNA talk page). 188.66.35.217 (talk) 08:24, 22 March 2023 (UTC)[reply]

TechPowerUp's chip db is a very unreliable source evidently

[edit]

I'm leaving this note after catching yet another wrong info in their database: first it was Navi 31 (https://en.m.wikipedia.org/wiki/Talk:RDNA_(microarchitecture)#188.3_Mtr/mm²_for_Navi_31_looks_fishy), then Valve's Steam Deck (https://en.m.wikipedia.org/wiki/Special:MobileDiff/1142658488) and now your (otherwise very good) recent edit I have just corrected after a bit of research (https://en.m.wikipedia.org/wiki/Special:MobileDiff/1146350988) – all these errors caught in the course of a few months *entirely by chance*. So please don't rely on TechPowerUp anymore to quote chip data without cross-checking against other web resources and manufacturer's data. 188.66.32.163 (talk) 10:43, 24 March 2023 (UTC)[reply]

Indeed, you are actually right! That's also what I've noticed about TPU's database lately.
The Ryzen 5 4600HS entry incorrectly states 11MB for L3 cache, that 11MB is actually L2+L3 combined total, and 8MB is the actual amount of L3 cache in that chip.
The Radeon 760M entry also completely incorrectly states 6 Compute Units, contrary to AMD's own site which states 8 Compute Units. Notebookcheck's own database entry seems spot on as far as specs of various mobile Radeon integrated GPUs go, although unfortunately the site isn't nearly as detailed and comprehensive as TPU (it doesn't have number of TMUs and ROPs, for example).
I have sent an email to the person who manages the database via the "Report an error" button for the Ryzen 5 4600HS entry, regarding the incorrect cache size, but it's been over 3 months since I did no and still no response, let alone the correction.
Uh, I wish I could respond to you on your own talk page instead of on Rando717's, but it seems your IP address changes around a lot, which makes it difficult to sustain a discussion on one user talk page. One of the great advantages of editing with an account as opposed to editing anonymously is you can sustain discussions on your user talk page even when your IP changes. Not to mention your contribution history won't be potentially mixed up with other people's edits on the same IP address(es).
Regards, AP 499D25 (talk) 02:11, 5 April 2023 (UTC)[reply]

Changing date format in references

[edit]

It's actually not necessary to "correct" the formatting of dates in all the citations in an article to be consistent with the other dates used in the article.

The reason for this, is that there is very likely going to be a "Use ___ dates" template at the top, which automatically changes the date format in citations to what's specified in the template (dmy or mdy). So for example if it's set to MDY dates, then regardless if it's ISO, DMY, or anything not MDY, it will show up as MDY to the reader.

If there isn't a 'Use mdy dates' / 'Use dmy dates' template at the top, then add one yourself.

P.s. I used to "fix" dates in references too, until I found out about this trick with "use ___ dates" templates. I just find it a waste of time now, along with "unredirecting" links too. Wikipedia:If it ain't broke, don't fix it.

You'll often see me insert DMY or ISO dates in references these days, if it's DMY then that means I used the insert cite template tool in the source editor toolbar, if it's ISO then I used the VisualEditor to do so (these seem to be the default date formats for each editing tool respectively).

Regards, —  AP 499D25  (talk) 00:10, 13 April 2023 (UTC)[reply]

@AP 499D25
I am aware of "use dates" templates... and I also used to change entire articles (refs, redirects, etc). :D
These days I do it only on sections as a side edit. It looks cleaner while editing wikitext, seeing for example 2020-04-02 is confusing at first. I know it's April, but it could be a typo.
I do the same with template redirects (shortcuts), for example changing from NA to N/a. As a side edit it doesn't take too much time. italic templates are triggering my imaginary OCD when checking "templates used in this preview".
I stopped doing this type of edits, aka "fixes" to references or redirects only.

Also when editing only a section, date template doesn't apply in previews. Rando717 (talk) 01:31, 13 April 2023 (UTC)[reply]
Makes sense, got it 👍.
Some advantages of using no pipe links are that it takes up less code / space in the source code of the article, and if that redirect becomes an article, the link takes you to that article straight away, you wouldn't have to go and update it to link to the article.
—  AP 499D25  (talk) 02:38, 13 April 2023 (UTC)[reply]

Epyc CPU lists in sandbox

[edit]

Regarding the "dilemma" of the Epyc CPU tables redesign in your sandbox:

  • Q1: Yes, I support merging of cells with same data
  • Q2: L3 cache size should be total, however what could be done regarding L3 per CCX, is write it under the "common features" list at top, saying "L3 cache: shared 8MB per CCX" or something like that.
  • Q3: No. Makes things less clearer in my opinion, a reader might get carried away by lack of I/OD in the CPU at all.
  • Q4: Are they part of the normal Epyc series (e.g. Epyc 7002)? If so, yes.
  • Q5: Probably yes, given that Intel CPU articles have it also, and no problem.

I've also noticed the Epyc article has all the full CPU list tables in them while there does not exist a "List of AMD Epyc processors" article, I think the tables on the Epyc article ought to be removed and replaced with summaries of the CPUs and description about each generation, while the said list article should have all the tables with each individual CPU and all the info about them. This is how the Opteron and List of AMD Opteron processors articles are arranged.

Hope this helps.

Regards, — AP 499D25 (talk) 12:53, 20 April 2023 (UTC)[reply]

Thanks for sharing your input. It helps.
About Q4: There are OEM models as part of series (e.g. Epyc 7002). But there are no official info about them on AMD site, same with CSP (Cloud Service Providers) models.
There are Amazon Web Service, Google Cloud, Microsoft Azure and Oracle Cloud Interface exclusive models. Similar to Xeon Apple exclusive "OEM" models.
Only problem is finding solid references about OEM models (not CSP). We can't use ebay or similar sites as a source.

I agree, there should be a "List of AMD Epyc processors" and all tables should move there. Rando717 (talk) 09:45, 21 April 2023 (UTC)[reply]
Response to Q4:
I wouldn't worry about it too much, for now.
I suppose there's two ways we can go about it.
One is to leave them out of the table / not include them for the time being, and maybe in the future, sources will pop up, or another editor might decide to add it.
The other is to simply include them all in the table, even without citations. As long as it's something you can google and find the specs of it easily somewhere. What actually matters here is verifiability, you can take the info from the table and compare them to other sources and they will be identical data / verifiable info.
As for creating list of Epyc article and moving tables there, that's yet another one of many tasks recently added to my to-do list. — AP 499D25 (talk) 12:58, 21 April 2023 (UTC)[reply]

New message from AP 499D25

[edit]

 You are invited to join the discussion at Talk:List of Intel Core processors § Table redesign. — AP 499D25 (talk) 05:39, 8 May 2023 (UTC)[reply]

Happy New Year, Rando717!

[edit]

   Send New Year cheer by adding {{subst:Happy New Year fireworks}} to user talk pages.

New AMD hardware announcements Computex 2024

[edit]

Hi Rando717,

I just thought I'd let you know about what I have, and haven't, written/covered about all the many different things shown at the AMD@Computex keynote today.

So far, I have created/added:

  • Ryzen 9000 series desktop processors list
  • Ryzen AI 300 series mobile processors list (such a dumb move copying intel's naming system I know!)
  • Some more info on Zen 5, including sneak peek of Turin server CPUs
  • Info on new processors on the Ryzen article
  • Ryzen 5000XT processors (damn they still at it making more CPUs for AM4 seven years later)
  • Radeon Pro W7900 dual-slot

What I haven't written about yet, are:

  • AMD XDNA (I couldn't find secondary independent sources that cover XDNA in detail sadly, at this time, so it won't meet the notability guidelines for now)
  • Info about new Epyc CPUs on the Epyc page (also note, they released 4004 series few weeks earlier but that hasn't been added either)
  • RDNA 3.5 (I have no idea what this is, I looked it up and I couldn't find any sources talking about what it is either).
  • New X870E and X870 chipsets (I think we would need to revise the table layout a bit, e.g. add some columns, so the reader could really see at a glance what the difference between them and the 600-series chipsets are)
  • Mention of extended longevity of the AM5 socket to "2027+" (source, not sure how to incorporate this into the article text)

So yeah, at this point in time the Epyc page isn't updated and links to RDNA3.5 and XDNA / XDNA2 are red links. I suggest leaving them as red links (i.e. not piping them to make them a blue link) until someone creates the articles or redirects them to coverage of it in a related article (e.g. an RDNA 3.5 section on RDNA 3 page).

That's it for me today, I have spent a lot of time and effort publishing all this new stuff, I'm a bit exhausted now. Some unusual editing choices/decisions I made may be pointed out in the edit summaries that I left.

By the way someone split up release date and pricing on Template:AMD Radeon Pro W7000 AGAIN (you probably know who I'm talking about), so I have kind-of undone that but some edits were kept, it's kind of a mess. — AP 499D25 (talk) 08:46, 3 June 2024 (UTC)[reply]