Insider look at Nokia design – how Nokia made money from innovation and design

| August 12, 2013 | 28 Replies

inside

An old document has popped up that looks at some Nokia Designs (I’m guessing for internal purposes only)

http://www.filedropper.com/nokiarm645testing

This is discussed over at TheVerge forums by “*****” an ex Nokian who had these documents at the time before said devices were not even announced. They note that it’s probably a good thing they don’t have schematics of future Nokia devices, like the rumoured phablet, due to Nokia’s tight security.

In the link below we’re informed how Nokia’s designs helped them make phones quickly, cheaply and thus so much money.

http://www.theverge.com/2013/8/11/4612604/an-insider-look-inside-nokias-designs

We hear frustrations from this Nokian of Nokia reusing old parts because it was cheap or balanced the books OR had spare parts (hello N97!)

On the sound quality part, they note Nokia pays quite a lot of attention to the audio capabilities of phones. There’s a bit of camera talk (oh edof) but not too interesting here for ‘basic’ phones. We’re also told of Nokia’s USB-OTG implementation of which droids didn’t get till 4.0 and many apparently still not matching the USB capabilities of rm-645.

Nokia apparently had a ‘tick tock’ schedule. Tick to introduce innovation, tock to produce simplified versions with minor improvements (making use of the initial R&D)

Cheers “*****” for the tip!

Category: Nokia

About the Author ()

Hey, thanks for reading my post. My name is Jay and I'm a medical student at the University of Manchester. When I can, I blog here at mynokiablog.com and tweet now and again @jaymontano. We also have a twitter and facebook accounts @mynokiablog and  Facebook.com/mynokiablog. Check out the tips, guides and rules for commenting >>click<< Contact us at tips(@)mynokiablog.com or email me directly on jay[at]mynokiablog.com
  • krishna6233

    n now nokia have wp!

    • Paul

      It is still possible to see some “tick and tock”.

      And a quote:
      “Finally, look at this:
      S60 Binary Compatibility
      SDK API
      Needs to be tested on every production release.

      Yeah, s60 had quite a few compatibility issues between the different versions and vendors. Extensive testing was needed to insure just basic compatibility. I obviously do not wish to return to the bad old days of symbian software development *shudders*.”

  • richard

    And that’s what we call business.

  • SLAYER

    I can’t remember the Rm 424, does anyone know what that was?

    • muhs

      noki 6720

  • http://twitter.com/haranguemnb Harangue

    The explanation as to why Nokia is the spot they are in now is that exact forum post. I can’t recall how many phones Nokia made that used the very same ARM1176 core design. High-end, mid-end even low-end at a later time all used the same chip.

    Rember the N97? If rumours are to be believed it had the very same internals (as in exactly) as all the early S60v5 phones. All 434MHZ ARM1176 and similar memory. That’s why nearly all the early touch devices were haunted by the lack of sufficient C: memory.

    The cost saving measures Nokia took made the books look rosey and allowed them to sell in great numbers because the devices were so cheap to make and thus could be sold cheap as well. Yet they did sell at a premium. Didn’t the N97 initially cost something like 500-600 euros?

    By stockpiling similar parts Nokia effectively cut themselves out of the market of yearly HW upgrades that were becoming the norm just about the time Nokia launched their first touchscreen devices.

    Now this is highly hypothetical, but Nokia isn’t making as much money or selling as many devices now because they can’t sell them with a high margin or dirtcheap like they could in the past. Nokia put themselves in a businessmodel that was bound to come to an end. We all know what happened.

    • jiipee

      That tells a lot about the mistakes they did in the accounting department. They were unable to include the cost of software development work done to match the underpowered hw as well as the delays.

    • Random Random

      Yes.

      Also, by making underpowered hardware they made it impossible to have future Symbian updates for the older devices.

      That’s why effectively the Symbian installed base for Symbian^3 was just 5 million in the end of 2010. Both Android and iOS had a much larger relevant installed base.

      That was also one reason why there was no migration path from Symbian to MeeGo. In reality the actual relevant Symbian installed base was almost non existent. In 2010 Nokia had maybe a 5% market share for phones with a modern touch screen.

      No wonder Nokia had to capitulate in 112. The just was no future for Symbian because the relevant installed base was 5 million in the end of 2010.

      • jiipee

        I dont get your logic. Didnt the underpowered hw mean that they could not update the older installed base to newer releases?

        There is some merit to your reasoning on the istalled base. Id like to hear from people knowledgeable on the Qt adoption at Nokia what the status actually was.

        • Random Random

          Yes.

          Underpowered hardware made it impossible to use Symbian^3 on the older hardware.

          There were several issues with Qt.

          - it was really really late
          - installing it on the old hardware was not that easy
          - supporting different resolutions and input methods was not easy enough.
          - On older hardware Qt was taking too much memory
          - how to distribute Qt applications for the older hardware?

          In the end of the day even N8 didn’t initially ship with a robust enough version of Qt.

          • jiipee

            “- installing it on the old hardware was not that easy”

            That was true before 2011. At least a friend of mine working for Digia claims that there were major improvements in early 2011 in that regard and also related to S40.

            I’ve heard about the initial Qt version for both ^3 and Meego and they werent fully finalized. Also the Qt5 will bring changes. Still, I havent heard that those were critical to the success.

            • Random Random

              No, there was no real improvement on that area. Just some technical advances.

              I’ll give you an example. A guy has Symbian phone and just like most of the people, it’s not updated and it doesn’t have a proper application store.

              Because of the lack of the store applications are not bought.

              Now how was Nokia to put Qt to those phones? To most of the phones.

              And even if they managed in that, there were lots of additional issues.

              • jiipee

                My friends 5800 received updates. Why couldnt they have pushed store via such updates? I dont say it was easy, but I dont see it impossible. The performance would have surely be more of an issue, but mentioning lack of store application does not sound like viable argument against it.

    • *****

      I don’t have the designs to confirm it, but I’m pretty sure (as in 99%) that the ARM 1176 core is STILL used by Nokia today!

      It is like the only ARM 11 core that can go up to 1ghz +, and so, it is probably what the Asha 300 line and Asha 500 line is using right now.

      Nokia was really successful back then with this business model. But really, once the spec wars started to heat up, Nokia really can’t keep up any more. (release phones in 2008 – 2010 running the 1176 clocked at 480mhz, when the fabs get better manufacturing processes, clock the same core at 680 in 2010-2011, then get a better SOC design, and clock the same core at 1ghz – 1.3 ghz in 2012 – 2013 and beyond).

      This is why I always supported the switch to Windows Phone. Too much of Symbian’s optimizations and drivers were designed with outdated chipsets like this in mind.

      • MdN

        And, guess who was making those chips that Nokia “had” to use?
        http://www.ifixit.com/Teardown/Nokia+N8+Teardown/3641/2

        • http://twitter.com/haranguemnb Harangue

          And where do you think the AMOLED panels they use now in the 925 and 1020 come from?

      • jiipee

        “This is why I always supported the switch to Windows Phone. Too much of Symbian’s optimizations and drivers were designed with outdated chipsets like this in mind.”

        That is exactly why they shouldnt have done WP, but Android. They still dont have anything but Qualcomm in use. They would urgently need Mediated etc support to get lower in price.

      • http://twitter.com/haranguemnb Harangue

        Yep, the ARM1176 is the only ARM11 design that can do 1GHZ(+) and can even be cinfigured in a dual-core set up AFAIK. I’m also inclined to believe that the single core 1GHZ phones Nokia still sells today are powered by the same chips that went in Symbian phones.

        From what I remember the first phone using the ARM1176 design was the 5800 (although it could have also been in phones prior to that) In the 5800 it was clocked at 343MHZ intitially, that went up to 434MHZ via an update. Later Symbian 3 launched with the same chip only to be clocked at 680MHZ, which was then hightened to 1GHZ in later designs like the 808.

        I don’t think Nokia ever went with chips that were baked at a smaller process, right? So, same design just smaller and inherently more energy efficient.

        As you rightly point out, Nokia really lost out when HTC and Samsung started with the Desire and Galaxy S respectively clocked at 1GHZ with an Cortex A8 design. That was 2010, and now we have quad-cores, 3 years later. It sure goes fast, a pace that Nokia could never keep up with if they kept going the way they did.

        From where I stand, I don’t think Nokia could have competed in that arena on their own. Although one can argue that they are now with the near Qualcomm exclusive on WP.

  • http://www.twitter.com/nabkawe5 Moody Caplan

    WOW, I kind of called it in my article (Microsoft fights old mistakes) I deduced how Nokia makes money !!!

    OK I’ll take a moment to gloat :D

  • *****

    Thanks for posting my tip!

    I’m actually not ex-nokia (or any part of the mobile industry for that matter). I’m a hacker, more in the “cook up custom roms” sense, and less “break into your computer”. I had a buddy who passed me loads of these documents a while back, from many different cellphone companies, on many different aspects like training, design, testing, accounting etc. I do not know how he got so many insider designs (sometimes even before a phone was launched). I got rid of quite a bit of them, as I didn’t want any trouble.

    Interestingly, on some design documents, you can see the designers themselves revising their own designs based on what parts Nokia had left over. In my opinion, it was Nokia’s cost cutting that really doomed Symbian(and I have the documents to back it up).

    Before 2009, the Japanese Symbian manufacturers always compared their wares to Nokia’s. In fact, when they were comparing their own designs with the legendary N-series (before the n97), they often admitted themselves that they “lost” to Nokia, and thus cannot hope to command the same price on the market.

    After 2009, Fujitsu was so confident in themselves, that they would say “Nokia is so far behind, they cannot hope to match our designs”. This was back in early 2010, when the C5-00 launched, Fujitsu was making high specced Symbian devices.

    In fact, I was able to get my hands some marketing materials for this phone: http://www.fujitsu.com/global/news/pr/archives/month/2011/20110721-01.html

    It is arguably the highest specced Symbian phone ever. Fujitsu themselves focused on promoting their “superiority over competing Symbian business devices like the Nokia E7″.

    Well, enough of this history. I’m pestering my friend to see if he can (hopefully legally) get me designs for the Nokia Lumia 1020 right now.

    • jiipee

      I think I saw this device in some articles back in 2011. It is interesting that it had two separate socs and that it was using OMAP for Symbian. The given 600hrs stand-by time on smartphone use sounds impressive.

      http://www.geek.com/mobile/fujitsu-f-07c-is-a-windows-7-pc-and-a-smartphone-1384465/

      • http://twitter.com/haranguemnb Harangue

        Well, the N95 (and N96 I believe) ran on OMAP chips as well. Not really odd that there was support then.

    • Random Random

      Yes.

      Cutting the costs.

      That’s how Nokia rendered the Symbian installed base irrelevant. They had so low quality hardware in Symbian phones that updating the installed base to Symbian^3 was impossible in 2010.

      That’s why Nokia’s installed base for Symbian was only a puny 5 million in the end of 2010. That was just a fraction of what Android or iOS had at the time.

      • jiipee

        As much I understand your reasoning, but why have you chosen end of 2010 as the point in time. If you were making logical reasoning, you should consider the installed base, when Nokia would reach the WP markets in volumes. Best would be to mention several points in time. You should at least mention the (estimated) status end 2011 without any change to strategy so that you could put yourself in the position of a decision-maker in late 2010. Something that helps to understand the final decision then was that Nokia had just started to redesing Meego UX only months ago.

        • Random Random

          The end of 2010 is a very good point in time to evaluate Nokia’s options.

          1) Approximately at that point of time Android exceeded Symbian in unit sales.

          2) In H2 2010 Android had surpassed Nokia’s Ovi Store in revenues and the trend continued strong.

          3) Nokia started shipping the highly anticipated Symbian^3 OS in the end of Q3 2010 and the Christmas season proved that it really changed nothing. The market share continued to collapse in Q4 2010 just as fast as it did in Q3 2010.

          4) In Q4 2010 it was also possible to sell Qt based applications to handsets with pre-installed Qt. While Qt may not have been fully ready, Nokia saw that Qt was not flooding Symbian^3 with applications. There were only minor improvements what it comes to applications.

          5) In Q4 2010 Nokia had a chance to fight against Apple because of the Antennagate incident. Even that proved to be less than enough to help Nokia stopping the trend.

          6) In the end of 2010 Android was rapidly starting to hit the lowest price points Nokia was selling Symbian with. This was an extremely dangerous for Nokia.

          About the projections.

          It was pretty easy to project the future Symbian sales by looking at the development of the market share. It’s still a valid metric for projecting what was going to happen.

          http://static3.businessinsider.com/image/4ec2c5a6eab8ea105200001b/chart-of-the-day-android-is-totally-blowing-away-the-competition.jpg

          Losing 5 points of market share in a quarter was true in Q3 2010 and Q4 2010. It was also obviously true in Q1 2011 and that didn’t happen because of 112. The projection was horrible and the collapse of Symbian sales was inevitable according to the trend. Remember that there was nothing to stop the trend because Symbian^3 failed on that. Really, what would have changed the trend?

          Just count the projected unit sales with market share dropping 5 points every quarter.

          In conclusion.

          The end of 2010 is also a very good point of time because it happened just before the strategy change. There was enough time to evaluate the situation and the success of Symbian^3 failure and make a decision.

          The end of 2010 was also a good point of time to evaluate MeeGo. It was planned that the team would be able to release a new phone in 2010 and at the end of 2010 that goal was obviously missed. It was the time to evaluate MeeGo.

          It must also be understood that N9 was not the MeeGo device Nokia was planning to release. The actual device may not have been ready in 2011 and it would have been postponed to 2012. N9 was considerably less ambient Nokia had planned to.

    • http://twitter.com/haranguemnb Harangue

      This line: ‘you can see the designers themselves revising their own designs based on what parts Nokia had left over.’

      That is corroborated by a few articles written about and with former Nokia engineers. It’s also stuff I heard about 2-3 years ago as major gripes from Nokia engineers and especially those working on next-gen HW or SW. They couldn’t just use new parts as they had to work with old stock first.

  • http://wgsupport.willowgarage.com/entries/50254887-How-To-Pee-pee-Your-Own-SOLAR-Might-Electrical-Energy-And-Twist-Turbines โคมไฟพลังงานแสงอาทิตย์

    This holds true for your pathways, walkways and wall lighs.
    For a classic western look, hand scrapped rawhide is much
    better than commercially processed rawhide. Simply
    cut the top of the bottle off at the desired height and polish the edges to prevent
    injury when drinking from your new glasses.

    Feel free to surf to my web blog; โคมไฟพลังงานแสงอาทิตย์

  • http://scotziyx.devhub.com find

    This is very interesting, You’re a very skilled
    blogger. I’ve joined your feed and look forward to seeking
    more of your wonderful post. Also, I’ve shared your site
    in my social networks!