Jump to content

Talk:List of Qualcomm Snapdragon systems on chips

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

Proposed renaming of new page

[edit]

Hey, I'd just like to propose a move of this new page to List of Qualcomm Snapdragon SoCs (instead of the current List of Qualcomm Snapdragon devices) or something similar to reduce potential confusion. This would pave the way for us to separate out the utilizing devices lists into a separate List of devices using Qualcomm Snapdragon SoCs page like was done with the List of devices using Mediatek SoCs and MediaTek pages. Charwinger21 (talk) 21:47, 6 July 2015 (UTC)[reply]

Sounds like a reasonable thing to do. --Guy Macon (talk) 23:48, 6 July 2015 (UTC)[reply]
WP:TITLEFORMAT suggests avoiding acronyms and abbreviations in titles. Is there something other than "SoCs" we can use to describe these devices? ~Kvng (talk) 01:26, 7 July 2015 (UTC)[reply]
Well, it only suggests against ambiguous abbreviations (not all abbreviations), but we could theoretically go with List of Qualcomm Snapdragon systems on a chip (or would List of Qualcomm Snapdragon system on a chips be better?) and List of devices using Qualcomm Snapdragon systems on a chip if we wanted to (kinda long in my opinion). I think ending it with "SoC" could be acceptable under WP:ACRONYMTITLE though (specifically under WP:ACRONYMTITLE#Acronyms as disambiguators), as the purpose is to disambiguate it from the main Qualcomm Snapdragon page, and we aren't likely to see any other page titled List of Qualcomm Snapdragon SoCs. Charwinger21 (talk) 06:33, 7 July 2015 (UTC)[reply]
SOC may be more ambiguous than you think. I agree that List of Qualcomm Snapdragon systems on a chip is long. Here are some other suggestions: List of Qualcomm Snapdragon chips, List of Qualcomm Snapdragon models, List of Qualcomm Snapdragon parts. ~Kvng (talk) 18:09, 7 July 2015 (UTC)[reply]
SOC by itself is ambiguous. SoC when qualifying Qualcomm Snapdragon SoCs isn't (which is the point of WP:ACRONYMTITLE#Acronyms as disambiguators). Some of those other suggestions could work (I'm partial to "chips" personally), but they get a little bit informal. IDK. Charwinger21 (talk) 23:49, 7 July 2015 (UTC)[reply]
@Guy Macon: any opinions? ~Kvng (talk) 22:03, 8 July 2015 (UTC)[reply]
Right now the page says that the Snapdragon 200 is a device, and 8225Q, 8625Q, 8210, 8610, 8312, 8212 and 8612 are models of the 200, so either models or devices are OK with me on that basis. Parts seems OK, chips seems too colloquial. SoC goes against WP:TITLEFORMAT and most people don't know what a SoC is, so I don't like it. If I had to choose I would pick systems on a chip -- who cares if it is a bit on the long side? --Guy Macon (talk) 02:36, 9 July 2015 (UTC)[reply]
If the article title is too long, then it may be hard to find. I think the current title is fine as-is.--Frmorrison (talk) 20:21, 15 July 2015 (UTC)[reply]
@Frmorrison: The reason for the name change proposal is an ambiguity problem with the current title. It could be read to refer to devices, such as smartphones, which use the SoC or it could refer to different models of the SoC itself. ~Kvng (talk) 20:55, 15 July 2015 (UTC)[reply]

New processors and naming

[edit]

From sources at codeaurora it seem likely that the next Snapdragon 835 is the MSM8998, codenamed MSM-cobalt. In addition there will be the Snapdragon X16 (4G) using the WTR5975 RF transceiver, and the Snapdragon X50 (5G) using the 28GHz SDR051 mmWave transceiver. Jahibadkaret (talk) 19:09, 14 December 2016 (UTC)[reply]


[edit]

Hello fellow Wikipedians,

I have just added archive links to one external link on List of Qualcomm Snapdragon devices. Please take a moment to review my edit. If necessary, add {{cbignore}} after the link to keep me from modifying it. Alternatively, you can add {{nobots|deny=InternetArchiveBot}} to keep me off the page altogether. I made the following changes:

When you have finished reviewing my changes, please set the checked parameter below to true to let others know.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

  • If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
  • If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—cyberbot IITalk to my owner:Online 20:46, 26 February 2016 (UTC)[reply]

[edit]

Hello fellow Wikipedians,

I have just modified 2 external links on List of Qualcomm Snapdragon systems-on-chip. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

  • If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
  • If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 05:44, 19 May 2017 (UTC)[reply]

[edit]

Hello fellow Wikipedians,

I have just modified 34 external links on List of Qualcomm Snapdragon systems-on-chip. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

  • If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
  • If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 02:25, 27 December 2017 (UTC)[reply]

Mass change and removal of entire page

[edit]

I had to revert the mass change and removal of information of the page as it made it a mess and hard to follow. Nothing on the talk page here shows this was discussed or gained consensus. ContentEditman (talk) 17:41, 28 January 2018 (UTC)[reply]

to User ContentEditman: All the information is in place. 1. Separated ARMv6 / ARMv7 / ARMv8, as on the pages of Exynos and MediaTek. This separates the old SoCs from the new ones. Now do not need to specify the architecture version in each line. All information is saved. 2. All tables get the same columns. Earlier, the 800 series had a different column structure, which was confusing, tables had exceeds the width of the screens. All information is saved. 3. The same information is combined. This is done according to the example of new models. For example, all Snapdragon S4 Pro chips support Bluetooth 4.0, 802.11n (2.4 / 5 GHz). Why write 3 times, if we can write once. All information is saved. "it made it a mess"? No. It has become easier to search and read. "hard to follow"? No. To switch from the modern 600 series to the modern 800 series, it is enough to press the PgDn one time. In the old version, you have to pass through all the obsolete SoC every time. — Preceding unsigned comment added by 91.207.170.192 (talk) 18:15, 28 January 2018 (UTC)[reply]

The information is not the same as a large amount was removed. The page was broken into groups that are not conducive to searching as it was before. You moved them to areas that are related to the ARM design, not the snapdragon design. Snapdragon is a SOC yet you tried to arrange them by CPU grouping. And you did this with no consensus or even one blip on the talk page. I edited the layout per Snapdragon design, not ARM. Please do not change till agreement and consensus is meet. ContentEditman (talk) 18:24, 28 January 2018 (UTC)[reply]
"The information is not the same as a large amount was removed." Give an example. If I accidentally deleted the information - it needs to be restored. "You moved them to areas that are related to the ARM design, not the snapdragon design." I did it on the example of a similar platform. This is an effective way to separate old chips from new ones. — Preceding unsigned comment added by 91.207.170.192 (talk) 18:51, 28 January 2018 (UTC)[reply]
Oh no, what are you doing !!!! You have removed a lot of small important changes!!!!!!!!! You have arranged sabotage!!!!!!!!!!!!!!!!!!! You do not respect the work of other people!!!!!!!!!!!!!!!!!!!!! How can you be so intolerant? — Preceding unsigned comment added by 91.207.170.192 (talk) 18:57, 28 January 2018 (UTC)[reply]
Just because it was done else where does not mean its proper here as the Snapdragon is a SOC and should be grouped by the SOC family, not the CPU which is a very small part of its design. Snapdragon is made up of many parts besides the CPU. By trying to group to one of its parts puts to much emphasis on that single piece. ContentEditman (talk) 18:59, 28 January 2018 (UTC)[reply]
If this is the only problem for you, why did you delete all other changes? In addition, a lot of work was done. Places could be rearranged. But you are lazy? Do you rightly ignore the work of others? It's insulting. You are a vivid example of tyranny in power. — Preceding unsigned comment added by 91.207.170.192 (talk) 19:08, 28 January 2018 (UTC)[reply]
And what "changes" are missing that you believe should added? Instead of calling people names; on Wikipedia you bring it to the TALK page and discuss it. So if there are changes you believe should be made please bring them up here and talk it out. ContentEditman (talk) 19:14, 28 January 2018 (UTC)[reply]
All the tables were format with one template. The headings of the tables were made the same. The Model number, the frequency, the number of processor cores, and the architecture were filled in by a single pattern. The hours of work were spent on this painstaking work.— Preceding unsigned comment added by 91.207.170.192 (talk) 19:22, 28 January 2018 (UTC)[reply]
What about this? Why did you delete all these changes? Just because I did not discuss it in advance?— Preceding unsigned comment added by 91.207.170.192 (talk) 19:57, 28 January 2018 (UTC)[reply]
@91.207.170.192
I am pretty sure that a lot of information about the new features of each SoC were lost with this whole revamp which is kind of sad. At least to me it was quite interesting reading about which SoC had which important changes. It was also a good measure of seeing how much an SoC improved over its predecessor.
Sadly a lot of this information was not written in the best way (a lot of grammar mistakes or just stupid writing), but the general information was definitely more than now, no doubt. 84.171.31.162 (talk) 00:10, 27 February 2024 (UTC)[reply]
@84.171.31.162
Nevermind what I wrote about the information being gone, I just found it, haha. Says a lot about information being structured in the wrong way though, why should the general specs of an SoC be in a different list/table than the information about the changes over the predecessor? This just feels quite unintuitive. 84.171.31.162 (talk) 00:13, 27 February 2024 (UTC)[reply]

If you do not like the division of the architecture of the processor, suggest how else to split old and new systems? This is not an archive. The information should be convenient. When between Snapdragon 670 and Snapdragon 835 two screens of old models it is inconvenient. — Preceding unsigned comment added by 91.207.170.192 (talk) 19:22, 28 January 2018 (UTC)[reply]

This is due to Qualcomms family naming of their SOCs. It is not up to wikipedia to break their family up by some small difference between them. If that was the case you could split them up by GPU design, modem, etc... Instead we follow their natural layout designed by the manufacture so it shows how the the manufacture differentiates them. In this case the numbering system they currently use shows what level of performance they are designed for and the market they are trying to capture. ContentEditman (talk) 19:29, 28 January 2018 (UTC)[reply]
You are approaching the systematization of information in terms of formal characteristics. This is ineffective. The page has increased too much - it's inconvenient to use. For example Snapdragon 600, 800, 801, 805 are archival models. They interfere with comparing information on modern models. Suggest a sign that you can divide the models into old and modern ones. On the pages of Exynos and MediaTek, used ARMv7 / ARMv8. This is a technical sign. Are you have another offer? You can suggest. But to keep between 670 and 835 all this is old (800, 801, 805) - it's uncomfortable.— Preceding unsigned comment added by 91.207.170.192 (talk) 19:48, 28 January 2018 (UTC)[reply]
Your examples, Exynos and MediaTek, do not design as much as Qualcomm does in their SOC. Snapdragon has its own custom GPU, Modem, etc... The others relay on others for their GPU and modem for example. So the CPU design is a larger difference in some of their chips so grouping like that for them may make sence in those cases. For snapdragon trying to group by ARM generation is highly flawed and puts to much emphasis on the CPU part that is only one very small part of the Snapdragon SOC. You keep looking to compare them but only looking at one small part. Thats like trying to group cars by their engine when that is just one small part. ContentEditman (talk) 19:55, 28 January 2018 (UTC)[reply]
Are you trying to read what I'm writing? The page has increased too much - it's inconvenient to use. Suggest a sign that you can divide the models into old and modern ones. Are you have another offer? But to keep between 670 and 835 all this is old (800, 801, 805) - it's uncomfortable.— Preceding unsigned comment added by 91.207.170.192 (talk) 20:02, 28 January 2018 (UTC)[reply]
Just because something is "uncomfortable" to you does not mean its not correct. You keep using that word for the reason you want to change it but that is not a good reason. Snapdragon is a SOC and should be broken down by the same way the manufacture of the chip and those that produce products that use it do so. You have not provided any good reason to change that other than emotion and personal taste. ContentEditman (talk) 20:08, 28 January 2018 (UTC)[reply]
Information is studied and consumed. It should be convenient to consume. Information should not only be beautifully framed. It should be useful. Most users compare modern SoCs. I realized that you can not understand this - we close this topic, I do not Don Quixote.
You can answer the question - why did you delete all other changes? All the tables were format with one template. The headings of the tables were made the same. The Model number, the frequency, the number of processor cores, and the architecture were filled in by a single pattern. The hours of work were spent on this painstaking work. — Preceding unsigned comment added by 91.207.170.192 (talk) 20:19, 28 January 2018 (UTC)[reply]
I added back the ISA part to the SOC that was removed. What is different now that you believe is incorrect? ContentEditman (talk) 20:36, 28 January 2018 (UTC)[reply]
Open the 2 windows. Compare. It's simple. A lot of small changes have been made to bring to a single table format. You destroyed all this. Your carelessness knows no bounds.91.207.170.192 (talk) 20:42, 28 January 2018 (UTC)[reply]
Аnd other changes that other users made. You destroyed all this.91.207.170.192 (talk) 21:12, 28 January 2018 (UTC)[reply]

RFC

[edit]

I'm breaking this here for clarity -- Cheers, Alfie. (Say Hi!) 22:48, 28 January 2018 (UTC)[reply]

Note: Previously, there was an RFC here. Now there is not. -- Cheers, Alfie. (Say Hi!) 00:33, 29 January 2018 (UTC)[reply]

It's not amazingly polite to revert a user who is being WP:BOLD, and even less polite to edit war with them. The whole point of that guideline is to encourage editors to make edits they feel there is consensus for. Then, if other editors disagree that there is consensus, they can discuss that on the talk page. Sure, the IP user is being somewhat inflammatory, but I'd really appreciate if you would listen to their comments; they are definitely trying to improve the content of the article: A someone who uses and develops on SoC-based computers (including some from Qualcomm), the architecture of the chip is a relevant way to tell them apart. I can't run an ARMv6 build of Debian on an ARMv7 SoC, even if everything else was identical. There is certainly some middle-ground here, and I think it'd be worthwhile trying to find that middle ground: I think you should both also read the MOS guidelines on Lists, if you haven't already. -- Cheers, Alfie. (Say Hi!) 22:48, 28 January 2018 (UTC)[reply]

Thank you for mentioning this topic. The operation of the other components of the SoC depends on the drivers. The version of the operating system depends on the version of the instruction set. For any SoC, always a key technical characteristic of the resolving compatibility with the operating system is a supported set of instructions.91.207.170.192 (talk) 23:41, 28 January 2018 (UTC)[reply]
And if this page was talking about CPUs you would have a point. But in this case this is a SOC and the CPU is a small part of that design. Even Qualcomm says Snapdragon processors aren’t ‘processors’ at all when talking about them. You keep focusing on one small part while the GPU and Modem designs are what differentiate the Snapdragon line from others. Trying to pigeon hole them by their CPU truly misses what makes the Snapdragon so successful and different. Qualcomm themselves even separate the Snapdragon by the number series here https://www.qualcomm.com/products/snapdragon/processors/comparison just the same way the page is built. ContentEditman (talk) 00:24, 29 January 2018 (UTC)[reply]
The operating system that runs the SoC runs on the CPU: everything else is a peripheral for the CPU. The "Number" of the chip isn't important to me if I'm not already familiar with Qualcomm's offerings, but I am maybe already aware of CPU architectures/instruction set versions, especially if I'm aware of other SoCs from competing manufacturers. -- Cheers, Alfie. (Say Hi!) 00:30, 29 January 2018 (UTC)[reply]
Yet the CPU is a very small part of the SOC. These are not CPUs but general SOCs. We already list what the ISA they run in their own column so its not like its hidden data or hard to find. The IP editor did being up that the page was to large/busy, something to that effect. I just put the Snapdragon S1 in a hidden/pull down setting. If its in agreement I can do a hidden/pull down on all those Discontinued Models leaving only the most recent and current models listed? Please see the Snapdragon S1 section now. If peopel thinks this looks worse I'll revert. Just trying to make every one happy, or less unhappy? ContentEditman (talk) 00:42, 29 January 2018 (UTC)[reply]
Earlier I tried to make the Discontinued Models hidden. I was attacked with a demand to return as it was. I was told that it's difficult for them to read in their mobile browser. I did not object. 91.207.170.192 (talk) 00:58, 29 January 2018 (UTC)[reply]
This conversation needs outside input from uninvolved editors, at this point. I'm far too tired right now to put together a proper RFC (it's 1:20AM!); For the time being, I'd like to see two or three options you can present to the RFC. I'd honestly be happy presenting these two revisions and asking users to pick between them, but if you both have actual proposals we could also put those forward. -- Cheers, Alfie. (Say Hi!) 01:23, 29 January 2018 (UTC)[reply]
I will leave the division of the ARMv7 and the ARMv8 to others. I have exhausted my arguments. I would like for other changes to be restored: Аll tables had a same format; The headings of the tables were made the same; The Model number, the frequency, the number of processor cores, and the architecture were filled in by a same format. All this was destroyed. Also other changes that other users made. 91.207.170.192 (talk) 01:35, 29 January 2018 (UTC)[reply]
What argument have you made other than I want it my way or I will just start name calling and throwing a tantrum as you have already done? I've tried to work with you but all you keep saying is all or nothing. ContentEditman (talk) 02:09, 29 January 2018 (UTC)[reply]

@ContentEditman: I'm calling you out here. I'm trying to find a solution to this dispute, and outbursts like that are not helpful. I understand this is frustrating, but, please: The IP has stated their proposal, please state yours. I'll come back to this in the morning an open an RFC if I have both proposals. -- Cheers, Alfie. (Say Hi!) 02:43, 29 January 2018 (UTC)[reply]

Thats not an outburst but a statement of fact. Him name calling and trowing a tantrum, as he has done several times above, shows he is not working in good faith. I already tried to reach out and work on some of his complaints but he does not want that. The Snapdragon article should be left as is, at least on the formating layout for SOCs. We should leave the Series as they are which is the same way Qualcomm list them on their site. Grouping them by their GPU, CPU, Modem, etc... is madness as any of those is one single item that makes up the SOC. ContentEditman (talk) 13:46, 29 January 2018 (UTC)[reply]
Return the rest of the changes. You accused me of deleting content. It's you who deleted the content. You can not mindlessly roll back the version a month ago. If you are not satisfied with the sorting of tables, please re-sort. Do not delete the rest of the content. Current version 18:21, 28 January 2018. Return it and re-sort the tables as you like. 87.226.160.250 (talk) 14:10, 29 January 2018 (UTC)[reply]
OK I'll ask again, what content is missing? I have asked several times yet all I get is insults. ContentEditman (talk) 14:19, 29 January 2018 (UTC)[reply]
How can you be so rude and so scoff at people? You are completely inadequate. If I have time, I will restore it.87.226.160.250 (talk) 14:31, 29 January 2018 (UTC)[reply]
So again all you do is throw insults and can't answer a single question of what has been removed. Also your last edit removed information for the APQ8064 including its reference. I will have to fix that now. ContentEditman (talk) 14:49, 29 January 2018 (UTC)[reply]
Let's calmly finish the editing. I'll check, I'll finish - then you can watch. You have deleted all the changes again. Stop it. Is a mockery.87.226.160.250 (talk) 14:53, 29 January 2018 (UTC)[reply]
Please stop your removal of well sited information. You are now edit warring and can be banned for it. ContentEditman (talk) 14:54, 29 January 2018 (UTC)[reply]
What information did I delete about the APQ8064? Information about Bluetooth has been moved to a single column "Connectivity". Are you kidding me? Are you completely insane? 87.226.160.250 (talk) 14:57, 29 January 2018 (UTC)[reply]
Thats because its hard to tell what you are doing when you don;t use the TALK page. I reverted that changes as i can finally see what you were doing. ContentEditman (talk) 14:59, 29 January 2018 (UTC)[reply]
I have explained to you several times what I am doing: Аll tables had a same format; The headings of the tables were made the same; The Model number, the frequency, the number of processor cores, and the architecture were filled in by a same format. Be attentive. This is extremely unpleasant. 87.226.160.250 (talk) 15:20, 29 January 2018 (UTC)[reply]
For ARMv7 SoCs I brought the tables to a single format. For the 800 series, many deleted characters are old table headers. All information is saved. If there is no objection, I will do it later for ARMv8. 91.207.170.44 (talk) 19:10, 29 January 2018 (UTC)[reply]
I have no objection and looks fine. Thanks ContentEditman (talk) 19:25, 29 January 2018 (UTC)[reply]

Modems information

[edit]

For all tables that contain models of SoCs that have only LTE version (and version without modem) I suggest removing "supporting LTE FDD, LTE TDD, WCDMA (3C-HSDPA, DC-HSUPA), CDMA1x, EV-DO Rev. B, TD-SCDMA and GSM/EDGE". This information is redundant, repeated in each line and takes up a lot of space: Snapdragon 410 and up, 615 and up. 87.226.160.250 (talk) 09:40, 30 January 2018 (UTC)[reply]

I agree that does seem redundant and also parts of the modem can be turned off for certain markets/phones so going to in-depth seems over frivolous. ContentEditman (talk) 17:17, 30 January 2018 (UTC)[reply]
I agree. User:Abune (talk) 15:05, 1 February 2018 (UTC)[reply]

Mistakes

[edit]

On SD 700 Codec i think it's not 240fps on FHD and only 120FPS on HD. — Preceding unsigned comment added by 2A02:120B:7F9:3CF0:A98D:5457:65DB:819A (talk) 10:34, 24 September 2018 (UTC)[reply]

"Devices using it"

[edit]

Is this column really necessary? The information it contains is, IMO, not very useful and makes the tables almost impossible to navigate, especially for popular SoCs. Perhaps these lists can be split out into a separate table listing only the SoC, if not simply deleted? 175.45.116.69 (talk) 23:27, 4 October 2018 (UTC)[reply]

I am also not a fan of it as it can also depend on the market what SoC a phone has. I would vote to remove but best to give it some time so others can add their thoughts here. ContentEditman (talk) 11:54, 23 October 2018 (UTC)[reply]

I removed the "Devices using it" from the Snapdragon S1 section. So the rest would look similar to that if its removed. After doing so I believe it looks much better and removes something that is not needed or useful. ContentEditman (talk) 13:13, 26 October 2018 (UTC)[reply]

Just cleaned the Snapdragon S2 section. ContentEditman (talk) 03:01, 28 October 2018 (UTC)[reply]

Many editors add information only about devices. To delete information from the main tables, the information should be transferred to additional tables or to another page. --46.39.231.245 (talk) 13:46, 28 October 2018 (UTC)[reply]

Just because editors add something does not mean its the proper place to add it. As I said above much of the information is not even 100% correct as a phone may use 1 SoC in the US but another in a different country. This information would be best on a article about the phone so it can be written out what SoC it uses in certain regions. This article is about the SoCs, not about phones. ContentEditman (talk) 14:31, 28 October 2018 (UTC)[reply]
Not ready to agree with you. Sometimes it is necessary to find out all the models of devices that use the SoC. I will not venture into the discussion - for me it is not critical.--46.39.231.245 (talk) 15:22, 28 October 2018 (UTC)[reply]

I just finished removing them from the legacy chips, Snapdragon S1,2,3,4. ContentEditman (talk) 14:20, 29 October 2018 (UTC)[reply]

Removed the rest. Much better and information is not pressed to tightly now. ContentEditman (talk) 00:27, 11 November 2018 (UTC)[reply]
Maybe you should move those informations to another page in my point of view, I am sure about many users are using these informations. TSSAT (talk) 01:02, 12 November 2018 (UTC)[reply]
Yes I brought that up before. The chipsets and phones are not always correct due to differences depending on the market. This would be best at the page of the phone, not the SOC page. That is why I removed them from this page. It took up to much room for information that may not be correct or properly placed. ContentEditman (talk) 01:06, 12 November 2018 (UTC)[reply]
The second person writes you that this information is interesting and important, it is grouped by SoC. Regardless of the market. You ignore. In the main tables, this is unnecessary. Delete without transfer - wrong.46.39.231.112 (talk) 04:41, 12 November 2018 (UTC)[reply]
I was not the one that made the request, But I did give it time for others to add what they thought. I even posted to the rfc and it came and expired. So no it was not wrong. It was removed after giving more than enough time. If you like to add the information to specific phone pages that may fit better. But the SoC page is not. ContentEditman (talk) 20:41, 12 November 2018 (UTC)[reply]
You mean someone has to move those informations because of your deletion? It's nonsense. You can delete these, and you have a responsibility to move them to another page, am I right? TSSAT (talk) 08:47, 13 November 2018 (UTC)[reply]
Not sure why you think its my "responsibility". As I said this talk section has been posted for a while and I made mention of it on my edits. I also added the RFC so others could join as well. Its not my job or "responsibility" to be able to read minds, that is why we have the talk page. You are still welcome to edit pages for specific phones and add the information if you think its notable on those pages. ContentEditman (talk) 12:44, 13 November 2018 (UTC)[reply]

I am disappointed this information was outright removed without relocating it or being preserved in any way. Other users have expressed interest as it provided aggregate information on all phones using that SoC. This would not make sense on any single phone page as it has been argued. Also the argument that some phones use different SoCs depending on region only applies to a tiny fraction of the phones and does not render the whole section invalid. This can be easily fixed by marking said phones and providing additional information. That being said, I plan to reintroduce this information into the article. I am open to suggestions on location and format, as I can agree that the previous version might've looked cluttered.Wikisthesia (talk) 08:15, 6 December 2018 (UTC)[reply]

I concur but we should probably have a different article altogether because as it was, it was very cluttered. What about "Devices using/based on SnapDragon SoC"?Artem S. Tashkinov (talk) 16:59, 11 May 2019 (UTC)[reply]
I have no problem with that but others may object if there are not good references to it. Esp if said device uses more than 1 SoC. Many devices use different SoCs for different regions or versions. Its one of many reasons they should not be on this page due to differentiation. ContentEditman (talk) 17:09, 11 May 2019 (UTC)[reply]

Came here looking up phones using a particular chip just to see "devices using it" section was removed for no good reason. Now the article looks more like Qualcomm marketing material rather than a practical wealth of knowledge. 86.23.32.51 (talk) 04:27, 9 January 2019 (UTC)[reply]

Same here. I come to this page every now and then for this information only. Why someone would just delete without moving it is beyond me. I'm really dissapointed. -188.97.113.165 (talk) 23:42, 1 February 2019 (UTC)[reply]
Same as above. Didn't know the column is gonna be deleted until it got deleted. Tiberiusteng (talk) 17:19, 26 February 2019 (UTC)[reply]

As this was agreed to before the information varies from market to market. That and this page is about Snapdragon SOCs, not devices that use them. Do not add back as this went through RFC before and was agreed upon with edits and updates made here. If you want to add to a device page that would be the best place for this information. ContentEditman (talk) 16:59, 5 May 2019 (UTC)[reply]

You are a vandal and a scoundrel. This is the desire of 2 people: yours and 175.45.116.69. The rest of the people disagreed. Your deed is disgusting. 46.39.230.105 (talk) 18:02, 5 May 2019 (UTC)[reply]
Attacks are not allowed. If you can not edit in good faith then do not do so. As said I made a RFC request and received no objections. I made edits with breaks for others to review. All done to Wikipedia rules. ContentEditman (talk) 19:18, 5 May 2019 (UTC)[reply]
"no objections"? You are a hypocrite. You ignore all objections that do not coincide with your opinion. 46.39.230.105 (talk) 20:13, 5 May 2019 (UTC)[reply]

I've gone ahead and created Devices using Qualcomm Snapdragon systems-on-chip from an old revision of this article. People are welcome to edit it. Artem S. Tashkinov (talk) 18:14, 11 May 2019 (UTC)[reply]

Heads up, the similar List of devices using Mediatek SoCs was speedy deleted after a half decade with no feedback from any of the article contributors. https://en.wikipedia.org/wiki/Wikipedia:Articles_for_deletion/List_of_devices_using_Mediatek_SoCs 198.52.130.115 (talk) 01:56, 27 February 2020 (UTC)[reply]

Similar platforms

[edit]

This is an exhaustive and unhelpful list section. How are these architectures similar?? The title of this section should be 'Different platforms' or 'Competing platforms', and contain informative text, not a list. Sbalfour (talk) 00:10, 11 November 2018 (UTC)[reply]

I made it a Collapsible list for now so it can be worked on. How would you like to change it? Expand or just leave a smaller collapsed list? ContentEditman (talk) 00:26, 11 November 2018 (UTC)[reply]
This links for quick access to the product page, which is close to the goal. What was bothering you? Collapsible list does not make sense.46.39.231.112 (talk) 09:10, 11 November 2018 (UTC)[reply]

Memory Bandwidth and Buswidth SDM630 and SDM660

[edit]

The net seems wiped of this information and references are hard to find. So this is a request/challenge to shed some light on it. If you have a look at certain LPDDR4 chip datasheets from Micron, there seem to be all sorts of configuration, resp. chip organisation per dram package (2x 16bit, 4x16bit, 1x 32bit, 2x 32bit, 1x 64bit). Note that the LPDDR4 standard only defines pin layout per dram die, not per package, and says that dies are split internally into two "channels" each of 16bit width.

So, as a manufacturer, you may be able to connect dram to the SoC using a dual channel, 32 bit wide memory bus connection, connecting each channel to

  • a single dram package, that incorporates two dies, each of them with 2x16bit data bus (totalling 2x 2x16bit data buses for that package)
but caution: dram packages with two dies exist, that share a single 16bit data bus per 1st, and resp. 2nd channels to each of the dies (totalling 1x 2x16bit data bus per package)
  • a dram package, that incorporate a single die only (totalling 2 dram packages for the system, each exposing a 2x16 bit wide data bus)
The second option can be identified by locating two chips on the mainboard, but the first option cannot be identified visually, as you're unable to tell whether a) there's a single die in the chip, or two or four, and b) how their data buses connect to the board/cpu (the ball contacts are beneath the chip, not on the side, and often the chip label can not be read as either tape or heat spreaders connect to the top; and even if you have a model number, it is a matter of luck to find the corr. datasheet on the net).

Looking at the publicly available SoC datasheets for the SDM630 and SDM660, you currently only find a dram frequency and the information of a "dual channel" memory bus, but it is unclear, if each of these CPU memory busses is 16bits or 32bits wide. Looking at a secondary source, phonedb.net, there is information about memory bus width for these Qualcomm SoCs (stating 2x 16bits memory bus for the SDM630, but 2x 32bits memory bus for the SDM660). There is a single reason to doubt this information, namely Qualcomm, that states that SDM630 and SDM660 are pin compatible (which suggests the pin layout of the memory bus per channel and the number of channels being the same for both of these SoCs).

Doing exhaustive crawling of the web, I was unable to dig up pin layouts for both of these Qualcomm SoCs, so I hope somebody else may shed some light on this issue. --91.55.169.239 (talk) 02:00, 13 November 2018 (UTC)[reply]

If unsure, what exactly to look for, have a look at the detailed datasheet for the SD820, page 15, that has a pin layout of the memory used, names number of dies for the dram package, etc. It clearly shows that the dram package used has 4 x16bit wide data bus to connect to the SoC (either single channel 64bit, dual channel 32bit, or quad channel 16bit). To decide which configuration is used by the CPU/SoC, a look at the pin layout of the SoC can be taken and in case of SD820 it is dual channel 32bit, according to this document. --91.55.169.239 (talk) 02:29, 13 November 2018 (UTC)[reply]
Referencing Geekbench4 benchmarks for the Moto X4 driven by a SD630 and Moto Z2 Play driven by a SD626 and having a close look at the benchmarks test component Memory Bandwidth, it is hard to believe that SD630 offers two channels of 32bits width (unless only one of two channels is used in moto x4).
  • The SD626 theoretical maximum mem bandwidth according to this wiki article is
~7.5 GB/s, derived from 1x 32bit/8 * (933 * 2) / 1000
  • If it is assumed (official reference missing, as stated above) that SD630 offers two 32bit wide channels to populate, then the theoretical maximum bandwidth for those systems is
~21.3 GB/s, derived from 2x 32bit/8 * (1333 * 2) / 1000
  • This leads to the conclusion that SD630 devices should exhibit a memory bandwidth performance of factor ~2.8 (21.3/7.5) compared to SD626 devices (in the optimum case that both channels are populated, of course). The Geekbench4 benchmark results, however, are nowhere near this factor. Instead, they show about equal results in memory bandwidth test component.
  • Splitting this factor ~1.4 still suggests that a higher memory bandwidth should be measurable using SD630 devices compared to SD626 ones. Splitting may be interpreted as either
  • dual channel possible, but only single channel populated or
  • dual channel populated, but each channel only 16bits wide.
It is hard to make a definite verdict on SD630 capability from device application, since there is no guarantee that each device makes full use of those capabilities. This means, even if the SD630 devices mentioned indicate a narrow bus width, we cannot proof that SD630 is indeed incapable of larger bandwidth (in other application scenarios). There are too many unknowns on that path of reasoning. A pinout or datasheet delivers more confidence in those capabilities than trying to do the math backwards. --93.201.175.1 (talk) 01:01, 14 November 2018 (UTC)[reply]
Actually, if System-on-Chip paradigm is strictly interpreted as dram-cpu-unit, there should be no room for device manufacturers to vary dram configuration. Theorem: A given SoC always has a specific dram configuration. If this generally holds true, we may claim with confidence, that Moto X4 (and others) use the SD630 SoC with both channels populated and that poor memory bandwidth can indeed be backtracked to a bus width of 16bits per channel, with the theoretical maximum at
~10.66 GB/s, derived from 2x 16bit/8 * (1333 * 2) / 1000
This would partially explain why SD630 shows equal or worse results in mem bandwidth tests when compared to single channel SD626 (dual 16bit and single 32bit do both result in 32bits width overall), but still, a moderate gain of factor ~1.42 (10.66/7.5) should be seen. --91.55.164.242 (talk) 03:38, 14 November 2018 (UTC)[reply]
There is bus width and mem bandwidth information in the anandtech announcement to SD630 and SD660. I have incorporated this into the article. It is in accordance to the info on phonedb.net for these SoCs, but both sources lack back-references to an original, upstream source. The announcement spares information about SD636, but at least according to phonedb.net, memory configuration of SD630 and SD636 does not match, so I've taken out the rowspan wrt this. --91.55.164.242 (talk) 04:17, 14 November 2018 (UTC)[reply]
It should be noted that some comments to the anandtech article doubt the correctness of a 2x32bit wide bus for SD660, based on comparing Geekbench scores. They suggest the SD660 data bus width to be 2x16bit (i.e. 16bit data bus per channel, dual channel controller in the cpu) as well, i.e. as narrow as SD630. The anandtech article claim for the SD630 is not disputed in their comments. If it turns out, that SD660 also has a narrow data bus, it means that current information in anandtech and phonedb is incorrect and that we should research the claims given for SD636 as well. --84.135.122.139 (talk) 20:14, 14 November 2018 (UTC)[reply]
extremetech.com reported the same memory bandwidth and bus widths in 2017 as the anandtech article did, but again no back-references to upstream sources given. --84.135.122.139 (talk) 20:33, 14 November 2018 (UTC)[reply]
androidcentral.com also claims Dual-channel LPDDR4 at 1866MHz 29.9GB/s, i.e. 2x32bit wide bus for the SD660. So there is a bunch of sources for this claim, despite the rumors that this supposedly does not pair with the benchmarks observable using Geekbench. --84.135.122.139 (talk) 20:43, 14 November 2018 (UTC)[reply]
anandtech.com's announcement of the SD670 claims 2x 16-bit @ 1866MHz LPDDR4 14.9GB/s, but 2x 32-bit in the announcement of SD660 earlier, in 2017. So there are two semi-official sources that portray the uncertainty very well. I've added both bus widths and these references to the article table for quick access and for readers to see that there's no definite answer currently. If anyone has a better source, do not hesitate to edit and update. --93.201.166.8 (talk) 02:16, 18 November 2018 (UTC)[reply]

SD Serie 700: FHD 120fps but HD 240FPS ?

[edit]

Seem to be an issue there. HD is possible with only 120 FPS while FHD is possible with 240 FPS ? — Preceding unsigned comment added by 2A02:1203:ECBD:A300:1412:A1AD:47CC:9BCB (talk) 14:42, 4 December 2018 (UTC)[reply]

I looked at the reference and it says 1080p (FHD) is 120fps. I swapped them around. I did not see anything about HD (720) but usually its double FHD so I left it for now. ContentEditman (talk) 18:19, 7 December 2018 (UTC)[reply]

Devices using Qualcomm Snapdragon systems-on-chip

[edit]

I've gone ahead and created Devices using Qualcomm Snapdragon systems-on-chip from an old revision of this article. People are welcome to edit it. Artem S. Tashkinov (talk) 18:03, 11 May 2019 (UTC)[reply]

What does "D & E" mean?

[edit]

There are a lot of accronym in the lower sections of the article, like "D & E". I couldn't find their meaning in the article. Could someone clarify this and edit the article accordingly? Please. Thanks!

That is the "Hardware codec supported" section. So D refers to Decode and E refers to Encode. ContentEditman (talk) 12:27, 13 May 2019 (UTC)[reply]

Why do 835 and 845 are listed together?

[edit]

These SoCs are vastly different in almost all aspects. Artem S. Tashkinov (talk) 13:31, 3 June 2020 (UTC)[reply]

For ease of comparison, for example. Dividing into separate sections on one line is inefficient. 46.39.248.87 (talk) 10:16, 18 June 2020 (UTC)[reply]

Move Hardware codec supported to Qualcomm Hexagon article ?

[edit]

I propose to move Hardware codec section to Qualcomm Hexagon article, because it is the Hexagon that handles this and to make the article cleaner, Bensuperpc, — Preceding unsigned comment added by Bensuperpc (talk • contribs) 11:28, 7 September 2020 (UTC)[reply]

Gen 1 Processors (2022) redundant

[edit]

Snapdragon 4, 6, 8 Gen 1 (2022) processors are listed twice.

RfC on 9to5Google as a source

[edit]

There is an RfC at WP:RSN on reliability of 9to5Google as a source.[1] Only 3 opinions have been given in about 19 days. More would be appreciated. -- Yae4 (talk) 15:31, 27 December 2022 (UTC)[reply]

2022-12-18 Page Move

[edit]

CoolingGibbon, please don't move large pages without discussing on the talk page, especially when the page name has been previously discussed on the talk page. 198.52.130.108 (talk) 11:26, 5 January 2023 (UTC)[reply]

Structure of SoC lists is bia

[edit]

The current structure of the page has one entry each for all of the 700 and 800 Snapdragon SoC lists, but also one entry each for the Snapdragon 7g1 series, 7g2 series, 8g1 series etc. which is quite biased towards the SoCs before the new naming scheme. This seems quite unhelpful to me, when the SoCs with the new naming scheme are much less in quantity (and success :P) and span a much smaller time each. Like, why would you have one chapter for a series spanning roughly a year and one chapter for a series spanning 9 years? That's ridiculous structuring. I propose merging the entries in the chapter view (I'm not sure what it's called) to have one entry for all Snapdragon mobile 7 series SoCs and one entry for all mobile 8 series SoCs, since having one entry for each SoC is a little too much and impossible with the regroup of all 6xx, 7xx and 8xx processors into one list each. 84.171.31.162 (talk) 00:26, 27 February 2024 (UTC)[reply]

It used to be that the page was separated into different generations for each family of SoC's (say, 630, 636 and 660 would be put together in one table, since they are literally the same sdm660family internally). Last year however someone decided to put all of them in one table, which is awful for browsing for 90% of all SoC's on the page. Bobos808 (talk) 12:23, 17 April 2024 (UTC)[reply]
I had a look in the edit page and it appears most of these changes were done by one person circa May-July 2023, with no previous approval or talk page comment as far as I can tell. I'd be happy to re-arrange them based on generation/family as most of the info about what processor belongs to what is not very hard to find. Bobos808 (talk) 12:37, 17 April 2024 (UTC)[reply]
Hello I'm the person who rearranged all tables.
First of all sorry for not starting a talk/discussion before making such drastic changes, I was quite new to making edits on Wikipedia then.
To explain why I grouped all 200, 400, 600, 700, and 800 series chips into a single table for each is that it took up a lot of space on the article, increased the size of the page making load times longer, and the grouping of the SoCs didn't really make any sense to me (sometimes they were from the same family like Bobos808 mentions but sometimes they just launched the same year without sharing much in common). Other than the identical tables the chapters only included the phrase "The Snapdragon XXX launched on [...]" so I summarized all those sections into a single one for each class of chips (no information was deleted btw). I also thought people wouldn't care too much about the outdated chips which was apparently a wrong assumption.
Doing the same for the "Snapdragon X Gen Y" chips like 84.171.31.162 (talk) suggested makes the most sense to me as we can always use Crtl + F or the search function on mobile to find the correct SoCs while the article would be easier to navigate in my opinion. RM12 (talk) 10:23, 22 April 2024 (UTC)[reply]
The latter would be better because at least the Gen chips are labelled in order, so in a table they'd appear roughly in the order they launched. To tackle the issue of chip families, it would be good to compromise and separate all chips in one family in their own section of a table, one next to another, instead of having each family have a separate table. This way, there wouldn't be wasted space but you'd still have the extra info of what chips are internally related and how they compare against eachother. Bobos808 (talk) 21:29, 23 April 2024 (UTC)[reply]

7s gen 3 gpu

[edit]

on the comment of this edit i forgot to say that only registered geekbench users can access more information of the run. here's a part of the info, anyone with a geekbench account can verify. 

      "id": 34,
      "value": "Adreno (TM) 810",
      "ivalue": 0
    },
    {
      "id": 40,
      "value": "0x0",
      "ivalue": 0
    },
    {
      "id": 41,
      "value": "volcano-userdebug 14 UKQ1.240411.001 eng.lnxbui.20240502.091457 test-keys",
      "ivalue": 0
    },
    {
      "id": 42,
      "value": "test-keys",
      "ivalue": 0
    },

117.2.112.159 (talk) 11:03, 1 September 2024 (UTC)[reply]

Yeah I wasn't logged, I logged in now and can see it, even though Geekbench sometimes is not the most reliable source but its good enough for me. Alawadhi3000 (talk) 12:30, 1 September 2024 (UTC)[reply]
Retrieved from "https://en.wikipedia.org/w/index.php?title=Talk:List_of_Qualcomm_Snapdragon_systems_on_chips&oldid=1243434353"