PTCL's EV-DO service is not worth buying until they get their act together. (Image credit: PTCL)

What a difference 18 months make! Back in April 2009 I was singing praises for PTCL’s awesome new EVDO service (now called PTCL EVO). I wrote this piece and documented performance numbers to show how well the service was working. In case you don’t want to go through the original article, we were experiencing 2.28Mbps from Texas to Lahore (downlink). Then, in March 2010 I wrote a followup piece, titled “PTCL Evo EV-DO experiences slow down, but performance still decent“. I documented that there had been about a 15% performance drop since the publication of the original post, but speeds were still quite tolerable and the service was generally good. I concluded that second piece with the words:

“I certainly hope PTCL will remedy the performance issues and restore download speeds to what they were at launch. But in the interim, I continue to be a relatively satisfied user.”

Well, unfortunately PTCL continues to disappoint. In stead of improving their service, the EVO product has been reduced to near complete uselessness. There’s no point in even publishing the numbers anymore since they vary greatly, but in general, I see an 8X slowdown from launch speeds. We’re talking pathetic performance in the .3Mbps range!!

The irony is that this performance is not limited to the original 3.1Mbps EVO offering. It is equally bad with the 9.3Mbps EVO Nitro product! I invested in one of these Nitro devices and was utterly disappointed with the experience. I was initially told that being an early adopter the issues I was seeing were primarily related to cell towers not having been upgraded with new firmware. But after 90 days of purchasing the device I was still getting 300Kbps performance… our IT team pursued this issue vigorously with PTCL only to discover a couple of utterly surprising factoids. First, PTCL automatically cuts back available bandwidth if you’ve downloaded more than a particular amount of data (1Gbps? 2Gbps?). We don’t quite know the exact numbers since the PTCL has never bothered to announce them. All we know is that this magic lever exists and performance suffers due to it. Second, PTCL has drastically increased the EVO subscriber base without investing hardly anything in upgrading available bandwidth. While some performance penalties are expected on EV-DO wireless connections as the number of subscribers in a cell increases, the kind of slowdowns we’re seeing have more to do with uplink limitations from PTCL to the public internet.

Given the amount of money PTCL can make (and probably has been making) with this offering, and the incredible convenience it provides when it works properly, it is a tragedy to be reporting all the above. Someone at PTCL HQ clearly needs to be taken to task and smacked around a bit. If this is how the new privatized PTCL is behaving, then one wonders what the value of bringing in private management has been.

If my experiences are vastly different from yours, please share your comments with us. We will also reach out to the powers-that-be at PTCL and see if we can get an official comment on EVO’s ridiculous status quo.

Share this:

It’s been about 3 weeks since the last post. I’ve got emails from many regular readers wondering where the hell I’ve disappeared. Thanks for asking, it’s gratifying, but rest assured, I am very much here. I’ve been traveling for work a bit too frequently and while I haven’t had much time to write during the last few weeks, the plus side of travel is that you always collect a whole bunch of interesting stories to share. And that’s exactly what I’ve done. So expect to see some of that interwoven into the next few posts.

2010 is almost done… it just whizzed by. And it’s getting to be that time of year when I break out the crystal ball, darken the room, chant my magical incantations and predict the future. I was just glancing at how my predictions for 2010 fared, and while I haven’t done the formal grading yet, I don’t think I was too far off. I guess the most glaring miss was the surprising political longevity of El Presidente. But more about that another day.

Just to satisfy the curiosity of those who have patiently been waiting for the posts to resume, stay tuned for an update on the PTCL-EVO situation, which I’ve written about extensively. Also, the fate of Wateen, working with Turkey, how to increase Pakistan’s GDP by 10-20% in a year or two by doing just *one* thing and generating electricity cheaply in the rural areas. All this and more, including some thoughts on retro computing. Thanks for sticking around and interacting in the message areas. There’s going to be a lot to discuss in the days ahead. See you around!

Share this:

PTCL needs to pay attention to the reduced QoS of its Evo service (Image credit: PTCL)

EV-DO has been a huge success in Pakistan. Since its debut in Pakistan early last year, multiple providers have entered the market and adoption has certainly gone through the roof. I first wrote about PTCL’s EV-DO services in Pakistan back in April 2009 and in the year that has gone by since then, the number of EV-DO users on PTCL’s service has exploded. Consequently, we’ve been receiving many blog comments complaining about poor performance, connection issues and other troubles. All this while, I’ve been using PTCL’s service and have experienced fairly decent service, albeit with some sporadic issues. I decided to revisit the PTCL EV-DO performance question and conduct a comparison with the data rates we were seeing back in April 2009. We can all sense there’s been a slow-down, but by how much exactly?

We used the trusty Speedtest.net tools, just as we had a year ago, to conduct our bandwidth and latency tests. Here is what our test looks for a Lahore-New York city connection:

PTCL EV-DO performance 3/26/2010. Lahore to New York.

To refresh your memory, we saw download speeds of 1.57Mbps and upload rates of .23 Mbps in April 2009. So in comparison, the most noticeable reduction has been in the download department. Upload is close enough, with a .01 Mbps difference, to be a wash. Downloads, however, are 14% slower. This is unfortunate, because it indicates that PTCL has not upgraded international bandwidth at the same rate it has grown users. Obviously, if there’s more demand for the international circuits there will be some degradation in performance.

To some extent, however, slowdowns are to be expected when user bases grow rapidly. So, putting things in context, my personal conclusion is that despite the 14-15%  reduction in download speeds, PTCL’s EV-DO service continues to provide decent performance with good mobility. It’s still a good option for consumers that need mobile broadband. That said, I certainly hope PTCL will remedy the performance issues and restore download speeds to what they were at launch. But in the interim, I continue to be a relatively satisfied user. If things change, I will certainly let you know!

Share this:

The Universe continues to be random, but we try and extract some order from it here at TechLahore!

Ah, it’s that time of year again! And how quickly it comes. It seems like yesterday that I wrote the predictions reproduced below, but look, three hundred and sixty five days have passed already. We must continue the tradition we started in 2007 and built upon in 2008… so here we go!

I was right on most issues this past year, but wrong on a few. There are still a handful of days left in 2009 but my guess is nothing material will change, so I’ll bite the bullet and get into the annual prediction review…

Political Predictions:

1. Asif Zardari is heading down a dangerous path. He will be significantly weakened and potentially removed in 2009. The conflicts with Amin Fahim, Nawaz Sharif and now Yusuf Raza and Mumtaz Bhutto, are fundamentally unsustainable. Combined with the Benazir tragedy fading further and further into the past, time will rob him of his number 1 political ally: sympathy. Were he a sounder statesman, he would have used 2008 to build alliances; he has tried, but ineffectively. 2009 will not be good for him.Well, the “potential” removal didn’t happen, but has Zardari been weakened or what?! The authority to use strategic assets has been transferred to the PM, the run-in with parts of the establishment on the issue of the NRO left him embarrassed at best, the ISI apparently despises (and we don’t blame them!) the pretender, Hussain Haqqani, Zardari’s blue eyed boy, to the extent that visa applications are being double checked for security in Islamabad. Haqqani’s washington embassy is no longer trusted. The Long March was stopped only after Zardari back tracked and restored the suo-moto-happy CJ. And then there was the NRO disaster. One can go on and on. If Zardari resides in the Presidency it is only by the skin of his teeth. He has been weakened significantly, no question. 100%.
   
2. Despite rumours, there will be no martial law in 2009; the Army will certainly have their input, but they will abstain from direct political involvement.There were rumours. And there was no martial law. 100%.
   
3. In the US, Obama will be constrained in his ability to implement his election promises. In Iraq, he will find that he will either have to come to grips with an out-right US ‘loss’ in the sense the that government there, in 2-5 years, will not be pro-US. Or he will have to do an about turn on his election promises and find some way of prolonging US involvement. This will be hard to do, because the US cannot afford it politically and financially. Any progress claimed by the pro-surge camp, will be seen to be visibly undone – but we’ll get to that in 2010 and 2011.Also in the US, Obama will have a very hard time with Afghan policy. Troop levels are going up as we speak, but the US military will soon discover that even 70,000 soldiers in Afghanistan are probably 130,000 too few. Afghanistan will not be policeable directly by the US military and sometime in 2009 we will see the US buying off warlords selectively, claiming that everything is ok and trying to set a course for departure. Of course, underneath the surface, nothing would have changed and Afghanistan will revert to being Afghanistan.Obama campaigned essentially on an anti-war platform. He is still in Iraq and has increased his presence in Afghanistan. As I said a year ago, the US cannot afford this financially, but thanks to Bush, they are stuck. Blasts continued to rip through Iraq in 2009, including some of the biggest ones leaving hundreds dead in individual incidents. The buying off of warlords, including talks with the Taliban happened on various levels throughout 2009. I also said the US would set a course for departure from Afghanistan, and we now have a date in 2011-2012. So, I would say I was on the mark as far as this prediction goes. 100%
   
4. 2009 will not be a good year for Pakistan-India relations. While there is an understanding that two nuclear armed neighbours have to live in peace, there is also a realization in Pakistan that India is an ungracious and scheming neighbour that sees itself as a global power – which it is not. Due to these grandiose visions it has of itself, India feels it can deal with Pakistan brusquely, and in this it is deeply mistaken. The region will continue to pay the price, through 2009, of these Indian misconceptions.
   
   No surprises here. Whereas Pakistan was still keen to extend extra effort at the end of 2008, we finally got fed up of Indian intransigence and constant conspiracy and disengaged.
   

5. 

   The Naxalite Communist Insurgency has ravaged 170 districts in India

   Terrorism in India will continue to hit hard through 2009. The insurgency in Assam will not be controlled and will continue to get more serious. It already poses a threat to the Indian federation, and this threat will get more severe in 2009.

   The deadly insurgency in India spread from the state of Assam to more than half the territory of India. The map to the right, courtesy India’s own IBN, shows that the violence and rebellion have now spread to 170 Indian districts. 2009 was a disastrous year for security in India as hundreds of terrorist attacks occurred. The violence on the Afghan border has received a lot of attention in the international press, but the magnitude of the insurgency in India dwarfs the FATA situation. 100%

Financial and Economic Predictions:

1. There is a belief that the economic problems in America can be solved – to some degree – by spending gobs and gobs of money. This money is essentially being borrowed from the future and raised through excessive taxation. While in the long term, society in the US will pay the price on both counts, in 2009 at least, Obama will have to take measures to inflate the stock exchange – even if artificially. This will be done through the passage of trillions of dollars of bailout packages. Therefore, in 2009 the major US stock indices will likely see a very modest increase.Well, in spades on this… the stocks have bounced back over 15%. So you could say that while I was directionally correct, I underestimated the ‘bounce’ in the market. But do see below… 50%
   
2. Not to be confused with the increase in stock indices, the US economy on the whole will be in bad shape through 2009. Many major businesses will file bankruptcy, unemployment will continue to rise at least through the first quarter or two of 2009 and spectacular corporate failures will continue to make headlines for the foreseeable future.Most experts agree that with high unemployment, continuing corporate layoffs, jobless profitability achieved mainly through OPEX cuts, the US economy is far from healthy. There were high profile bankruptcies throughout 2009 and unemployment has been growing throughout the year. 100%.
   
3. In Pakistan, the KSE has taken a massive battering due to the incorrect policies of the SEC, KSE and the Ministry of Finance. Specificaly, the lower lock prevented investors from selling when they wanted to and that caused an avalanche when the lower lock was ultimately removed. But everyone realizes this. And 2009 will see Pakistani stocks head higher. I would guess by as much as 10%.KSE is higher by over 50%. I will give myself 50% here, because while I was directionally correct, the increase in Pakistani stocks has been far greater than I could estimate.
   
4. The trade gap in Pakistan will be reduced somewhat and the out-of-control imports that were a major issue through the second half of 2007 and most of 2008, will be arrested.This too, happened. Imports were cut even at the expense of a slower resolution to the power issues. Fiscal discipline has been fairly tight, but thankfully not at the expense of development projects such as the Ring Road, the reconstruction of Swat Valley, the signal free corridors in Karachi, various Dams and of course the Centaurus, ISE HQ and Telecom tower in Islamabad.
   
5. Food will be big on the international economic agenda and corporate farming in Pakistan will mark 2009 as the year of its birth. This is one of the biggest trends to watch in Pakistan and globally, for the foreseeable future.Things are going slow here. I will give myself zero % – there wasn’t much action this year on this front even though Mian Mansha continues to make investments in this area and several Arab states, including the UAE and Saudi Arabia have expressed serious interest and applied for Government of Pakistan permission to conduct corporate farming.
   
6. Iran will begin to supply about 1 Gigawatt of electricity to Pakistan and an additional 600-900MW will come online within the country in 2009. While this will reduce the power deficiencies to a great degree, it won’t be until 2010 that load shedding is firmly in our past.Meh… 50% on this one. The Iran electricity import did not hit 1 Gw though significant amounts are being imported to power many areas in Baluchistan. Mian Mansha’s Nishat power alone added over 250MW of capacity in 2009, Wind power plants close to 100MW came online in addition to some rental units. The Turkish ship-based power generation scheme didn’t materialize, but new capacity did approach that lower end 600MW estimate I had come up with. I expect more will happen on this front in 2010, but then more on that in another post.
   
7. The US$ will trade between Rs. 75 and 85 through 2009. While the Rs. might weaken, we believe the US $ isn’t on very firm ground either. One risk to this prediction is a potential devaluation of the Pakitani Rs. to spur exports and narrow the trade gap. We don’t think that would be a very smart move…100%. That’s exactly what happened.

IT Industry Predictions:

1. The Software Tower in Lahore will finally get wrapped up close to the end of 2009. Fingers crossed.I was wrong. This has gone into March of next year. 0%.
   
2. Netsol will remain in serious trouble through 2009 and will either face some sort of shareholder action or will become a dirt-cheap acquisition target.50%. Netsol remained in serious trouble throughout 2009, at one point their stock dropped to a paltry 25c on the NASDAQ. They’ve since recovered, but still under $1, are far from their 2008 peak of almost $3.50.
   
3. The economic downturn in the West will make it harder for Pakistani companies to expand services and we will start to see some very interesting Web 2.0 and other consumer-focused products emerging from Lahore, Islamabad and Karachi in particular. Due to the massive adoption of mobile technology in Pakistan, many of these products will cater to mobile phones/users.In spades, my friends, in spades. We have so many product based successes to celebrate this year… The recent article I did on Pepper.pk – a spinoff from a Pakistan based software services company, FiveRivers Technologies – is indicative of this trend. We’ve got Cricket Revolution from Mindstorm, with true international bestseller potential. Then we have Geniteam working on mobile apps. 2009 also saw a Pakistani product on the Top 10 list on the Apple Appstore… Jaadu VNC.
   
4. 2009 will be a good year for WiMAX in Pakistan. There are now two competing offerings and Wateen is working hard to resolve coverage issues. Pakistan will be a WiMAX success story.WiMAX has truly taken off in Pakistan. There are three providers and there is serious penetration. And the nice surprise has been the excellent EVDO mobile broadband services. See both our articles on PTCL/Wordcall’s services in this area. Beautiful! 100%.
   
5. Windows 7 will be released early – before the holiday season in the US – and it will do well. It will be the only new version of a Microsoft operating system that will have more modest hardware requirements as compared to the previous version, and this will be a key reason why it will succeed.Windows 7 is going to be a success. The reviews so far are pretty good. It already has almost 4% market share, and by any measure, it is on its way to being a blockbuster. 100%.
   
6. Android will threaten to become pretty big in 2009, but will not actually become a dominant platform for the next year or more.With the Motorola Droid riding high in Q4 2009 and Android devices ready to break through big time in 2010, I couldn’t have been more right on this one. Android doesn’t have threatening market share this year, but just wait and see.
   
7. Palm will come out with a ‘Hail Mary’ piece of technology during CES, and it will not work for them. Palm will be pretty close to an untimely death by the end of 2009.Are we ready to call the Pre a failure? I think so. It’s pretty clear there is no room beyond Android, iPhone, Windows Mobile, Nokia and RIM for a SIXTH platform. And we haven’t even counted Samsung’s new mobile OS and all sorts of proprietary solutions in use today. No sir. The Hail Mary piece of technology I predicted in 2008 turned out to be the Pre, and it is a failure.
   
8. PSEB will remain relatively ineffective through 2009; since Yusuf Husain’s departure, this has unfortunately been the case and will continue to be the case… The primary reason is that the Ministry of S&T is being fundamentally mishandled and with the government’s occupation with all sorts of ridiculous infighting, the PSEB is not getting the attention necessary to attract and retain a top-class Managing Director.I was polite about this last year, but let me just say it out loud now. Can we boot this useless bureaucrat Talib Hussain out of the PSEB already? What has he done? He is utterly unimpressive, he knows nothing about IT, he has absolutely no vision and he doesn’t belong in this department at all. Boot him out and replace him with someone who knows his bits from his bytes. Seriously. This guy is zero in technology, zero in marketing and zero in, well, pretty much everything else. Get him out and stop wasting our time. We have to develop this industry and he is not the man to do it.

So, let’s see now. Where do we stand? 20 predictions, 2 wrong and 4 at 50%. That’s 16 out of 20, or about 80%. There you have it folks. Get ready for the 2010 list!

Share this:

PTCL's EV-DO service is sweeeet!

The communication infrastructure in Pakistan, both wired and wireless, is quite incredible. There’s one pet pieve I’d had for a while though, and that’s the lack of inexpensive, true mobile broadband in Lahore. WiMAX is available, but I haven’t come across a USB dongle from Wateen so far. Their rates are also a bit on the higher side, with download caps and other nonsense. WiTribe and Mobilink are offering their services in Karachi only at this time. I’ve tried Telenor’s Edge USB dongle and PTCL’s WLL USB mobile. The former is too slow for full-fledged web surfing, video watching etc. and the latter – though a bit faster – is too clumsy. You end up having to plug-in a mobile WLL handset, a candybar phone, into your USB port. The thing hangs out and hampers mobility. I just don’t like it. And it delivers only about 20Kbps. Just not worth it.

So, imagine my surprise when I tried out PTCL’s new EV-DO service. I paid Rs. 8,000 for one-time connection and hardware purchase charges. I got a sleek looking, black EV-DO wireless USB dongle with an extensible antenna, that happens to go perfectly with my T60. The dongle has a built USB storage, so all I had to do was plug it in and the setup program started… from the dongle itself! Pretty cool. It took me about 5 minutes to run through the install, and boom! I was in business. My first stop was google.com, which loaded instantly. I could already see the difference in performance but I decided to measure it a little more objectively. Speedtest.net is always handy for that.

Speedtest detected that I was coming in from Lahore and using PTCL’s Triple Play network. It automatically selected Islamabad as the closest server site that I should run broadband tests against. But I wanted to be a little more adventurous – and realistic. I picked a few sites in the US since many of the websites I use are hosted there. That would be a true performance test, I thought to myself! And here’s some snapshots of the results I got.

Speedtest results for PTCL EV-DO

Speedtest results for PTCL EV-DO

Pretty insane! I’m getting 2.28Mbps downlink from Texas and 1.57Mbps from New York! I thought this was too good to be true so I reran the tests multiple times. The results were in the ballpark each time. Over the next couple of days I tried the tests at different hours of the day. No difference. These are real!

I did notice that the latency was somewhat higher than I would have liked, but unfortunately that is one of the issues with wireless systems like EV-DO. You pay a bit in the latency department for the bandwidth and mobility it gives you. Further study of the EV-DO spec does indicate that Rev 2 (or B?) optimizes latency somewhat. I am not sure whether PTCL has their transponders upgraded to this spec. It’s a software-only upgrade, so if they don’t already, I hope they do this soon.

The monthly charges for PTCL’s EV-DO service are a flat Rs. 2,000. Only $25! That’s half of what you pay Time Warner for fixed Roadrunner cable and about 1/4th of what you pay AT&T for an unlimited package on an AirCard. The value and performance are just amazing.

I’m a convert, if you can’t tell. I think this is PTCL’s best service offering since they launched WLL a few years ago. If you need to be mobile and require high-bandwidth access, PTCL’s EV-DO is indispensable. Get it today!

Share this: