Looking back at the response to the release of 1.02, a significant part of the discontents appeared to be the result of a poor explanation of how the various changes affected gameplay. A simple listing of all the change, as in the readme, can be insufficient to explain how the changes will affect the evolution of the game. As such, the betas decided to put together a summary of the most important changes to the 1.03 and their impact on gameplay, so that gamers would have a better idea of what to expect when they load their first game with 1.03. Note that the following is certainly not a comprehensive list of all or even most changes, but rather focuses on the most fundamental changes. Also please note that this is not an official document from Paradox, but rather the opinions of a group of betatesters.
Broadly speaking, there are three major changes to the economic model for 1.03. First, the starting factory setup for the Grand Campaign has been overhauled, with, for example, many more fabric factories, glass factories, etc., along with more accurately reflecting the historical level of industrialization in the 1830s. Second, the AI is much better at building the right types of factories, in terms of building a balanced economy and in terms of addressing shortages. Third, the base production efficiencies of certain goods (like fabric) have been tweaked to ensure a more balanced production of goods (also note that the base production efficiencies have been exported to a text file for easy modding). Combined, tests have shown that these changes result in a more steady supply of vital goods on the world market, including regular clothes, paper, etc. This does not mean that a player can always buy whatever goods he wants in whatever quantity he desires (a player's position in the prestige ranking does matter), but overall access to these types of goods has improved, not just for major powers, but for minors like Portugal and Serbia.
Partisans have been altered with the 1.03 patch, in that the frequency with which they appear has been halved, and they are slower in occupying provinces. Partisans also are now truly unreinforceable and are demobilised when the nation is completely at peace. The results of these changes is that partisans remain a concern, but usually not to the sometimes-overwhelming degree of 1.02. In terms of naval warfare, AI countries are much better at building advanced warships. For example, in a recent game a total of approximately 60 dreadnoughts were built by 1920 by four different countries. However, it should be noted that dreadnoughts, whilst common, are not built in every game due to the random nature of the invention process (the same applies for tanks). Also, AI countries are much better, but not perfect, at demobilising troops after a war.
The most important change to diplomacy has been the addition of much more detailed protect, defend, and combat sections to the AI files. This results in AI countries being much more active in diplomacy, particularly with respect to wars. Wars, and the ensuing transfer of territory, are more common in 1.03 than in 1.02, so consider yourself warned. Also, it should be noted that AI countries will no longer DOW countries when it would violate an existing peace treaty, which means that the endless BB wars will not occur, although they may declare war again in five years. One other major change is in the peace process. A white peace signed after there has been no combat between two sides for 24 months costs no prestige. In addition, white peaces between alliance leaders and junior alliance members, such as the UK's Indian satellites, no longer cost prestige. However, a separate white peace to get out of a war which your ally has called will cost prestige, if applicable. One word of warning - it has been found that making any kind of alliance with an enemy of an existing alliance partner whilst they are at war, even if it excludes the enemy of your ally, is likely to break your existing alliance. You have been warned! One other change that players might notice is that the USA and Switzerland are much less likely to become involved in alliances. Persuading these powers to ally with you will be long, expensive, and often futile.
There has been two important changes made to colonial buildings: a leadership cost of 3 has been added to each building, and the build time for each building has been extended to four years. The most important effects of these changes has been to slow the pace of colonization, although certain countries are better able to operate with these changes than others (for example, Britain's overall colonial performance is somewhat improved, while a country like the Netherlands no longer colonizes as much as in 1.02). Note that you will still see colonial buildings start to go up in Africa in the first year, but not as many and not as quickly as in 1.02. Also, it should be noted that the starting colonial building setup for the GC has been revised, so that there is no colonal region that can be claimed by any country at the start of the scenario.
A key change has been that a monarchy is more likely to change into a constitutional monarchy rather than become a democracy via reforms. Tests have shown that constitutional monarchies, not democracies, remain the most numerous government type (particularly amongst the larger powers) into the 20th century. At the same time, the constitutional monarchy government type has been made more appealing, as in a constitutional monarchy, plurality no longer increases militancy. As well, the appearance of socialist POPs has been tweaked. Generally speaking, it is only in the 1850s that even a small number of POPs become socialist, and it is not until the 1870s and 1880s that they become a major political force. Also, the effect of the tax rate on consciousness has been halved, while the effect of plurality on consciousness has been quartered. All told, these changes make a more conservative political strategy more viable and practical.
One important change to cultures has been the splitting up of the Spanish and Portuguese POPs of Latin and South America into seven separate cultures, which both reflects the diversity of the region and makes annexations of large amounts of territory by one country in that region somewhat more problematic. Also, only POPs under 10k will now merge, and POPs now split 75/25 as opposed to 50/50. As well, a large number of cases where there was significantly more population than historically was the case (such as Austria, Java, Tunis, etc.) have been fixed, which in particular has had the effect of reducing the overall strength of Austria. Finally, the resource demands of lower- and middle-class POPs have been reduced, meaning that it is more likely (but by no means guaranteed) that they will be able to afford the resources they desire (provided that said resources are available on the WM).
Some of the more important events have had major changes made to them. The Conservative Empire event (but not the other unification events) has been modded to reduce the amount of badboy incurred. We think it works as intended now, but the badboy command can be tweaked by players if they so desire (or applied to the other unification events). Also for Germany, the Three Hurrahs event has been changed to allow for the annexation of the SGF if the historical path is followed. As well, there is a new civilising event that is applicable to all nations, which is in addition to the hard-coded civilising options that Japan, Persia, and China get. China in particular has had its civilising event toned down, as it will now not get civilised status via historical events unless they choose an ahistorical option in 1899. Note, though, that they will still get extra techs and stuff if they choose the "learn from the west" option after the Opium/Arrow War defeats. Also, the Italian unification events have been made much easier to fulfill. It is no longer necessary to have the backing of France, and Naples rather than Messina is now the trigger province in terms of the Two Sicilies. Thus, it is now relatively straightforward for a player to achieve unification, and even the AI manages it sometimes. Finally, the USCA dissolution event is now much more likely to result in the collapse of the USCA rather than letting it remain, as was the case in 1.02, and the former USCA region is also much more volatile, with higher militancy and consciousness.